Cardiac CT imaging: diagnosis of cardiovascular disease [Third edition] 9783319282176, 9783319282190, 2016934867, 3319282174

Citation preview

Matthew J. Budoff Jerold S. Shinbane Editors

Cardiac CT Imaging Diagnosis of Cardiovascular Disease Third Edition

123

Cardiac CT Imaging

Matthew J. Budoff • Jerold S. Shinbane Editors

Cardiac CT Imaging Diagnosis of Cardiovascular Disease Third Edition

Editors Matthew J. Budoff Division of Cardiology Harbor-UCLA Medical Center Torrance, CA USA

Jerold S. Shinbane Keck School of Medicine University of Southern California Los Angeles, CA USA

Additional material to this book can be downloaded from http://extras.springer.com. ISBN 978-3-319-28217-6 ISBN 978-3-319-28219-0 DOI 10.1007/978-3-319-28219-0

(eBook)

Library of Congress Control Number: 2016934867 Springer Cham Heidelberg New York Dordrecht London © Springer International Publishing 2016 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. Printed on acid-free paper Springer International Publishing AG Switzerland is part of Springer Science+Business Media (www.springer.com)

Foreword to the Second Edition

Cardiac CT has finally come of age. After nearly 30 years of development and growth, tomographic X-ray is being embraced by cardiologists as a useful imaging technology. Thirty years ago, Doug Boyd envisioned a unique CT scanner that would have sufficient temporal resolution to permit motion-artifact-free images of the heart. In the late 1970s, I had the good fortune to work closely with Dr. Boyd, Marty Lipton, and Bob Herkens who had the vision to recognize the potential of CT imaging for the diagnosis of heart disease. In the early 1980s, when electron beam CT became available, others, including Mel Marcus (deceased), John Rumberger, Arthur Agatston, and Dave King (deceased), were instrumental in making clinical cardiologists aware of the potential of cardiac CT. In 1985, several investigators recognized the potential of cardiac CT for identifying and quantifying coronary artery calcium. Now, 20 years later, there is wide recognition of the value of coronary calcium quantification for the prediction of future coronary events in asymptomatic people. It has been a long and arduous road, but finally, wide-spread screening may significantly reduce the 150,000 sudden deaths and 300,000 myocardial infarctions that occur each year in the United States as the first symptom of heart disease In the late 1970s, it was thought that a 2.4-s scan time was very fast CT scanning. With the development of electron beam technology, scan times of 50 ms became possible, giving rise to terms such as fast CT, ultrafast CT, and RACAT (rapid acquisition computed axial tomography). Now, with the development of multidetector scanners capable of 64, 128, 256, and beyond simultaneous slices, spatial resolution is approaching that of conventional cineangiography, and the holy grail, noninvasive coronary arteriography, appears attainable. In this book, Drs. Matthew Budoff and Jerold Shinbane, preeminent leaders in the field of cardiac CT, have described the many and varied uses of the technology in the diagnosis of cardiovascular disease. The book clearly documents that cardiac CT has not only arrived but has become a very valuable and potent diagnostic tool. Bruce H. Brundage, MD, MACC Heart Institute of the Cascades, Bend, OR, USA UCLA School of Medicine, Los Angeles, CA, USA

v

Preface

It is a testament to the intellect and diligence of the physician-scientists and engineers involved in the field of cardiovascular computed tomography that a third edition of this text is necessary in so short a time since this field was created. Our first cardiac CT angiogram was performed in January 1995, over 20 years ago. The trials and tribulations that we faced over the introduction of this modality are reminiscent of a quote by the philosopher Arthur Schopenhauer: All truth passes through three stages. First, it is ridiculed. Second, it is violently opposed. Third, it is accepted as being self-evident. Cardiovascular CT has now matured into a firmly established subspecialty of radiology and cardiovascular medicine with a multidisciplinary society with 4000 members, a board examination, dedicated journals, numerous scientific statements from leading national societies, focused national and international meetings, and clear education and training pathways. The foundation has been provided by a sound medical literature, which continues to grow at an astounding pace. We hope that research maintains the same forward momentum fueled by the intellectual curiosity and passion to increase the understanding of the cardiovascular system and improve patient care. As we look ahead, we also continue to look back and acknowledge our debt to the pioneers of this technology who dedicated their careers to forwarding this discipline, who persevered the ridicule and violent opposition, and may or may not be enjoying the current acceptance and utilization in this “self-evident” era of cardiac CT. Cardiovascular CT has become a powerful risk stratifying tool for the early detection of atherosclerosis, used as a calcium score to identify asymptomatic persons at risk of CVD. It has also developed into the de facto noninvasive angiogram, a measure of coronary stenosis, a substitute for coronary angiography or noninvasive exercise testing in certain clinical situations, and a powerful tool to image the heart for congenital heart disease, trans-aortic valve procedures, perfusion imaging, and coronary anomalies. There has been a paradigm shift in its role related to cardiovascular therapies, with progress from pre- and postprocedure assessment to use in the actual guidance of a variety of invasive procedures. Advances in CT scanners, imaging techniques, postprocessing workstations, and interpretation for diagnostic and therapeutic applications have now made the field relevant to the entire spectrum of physicians who diagnose and treat cardiovascular disease. As such, a thorough knowledge of cardiovascular CT is required for the thoughtful and individualized application in patient care. We hope that this text will provide the substrate for a detailed understanding of the art and science of this technology. Torrance, CA, USA

Matthew J. Budoff, MD Jerold S. Shinbane, MD, FACC, FHRS, FSCCT

vii

Contents

Part I

Overview

1

Computed Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Matthew J. Budoff

2

Cardiovascular Computed Tomography: Current and Future Scanning System Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Rine Nakanishi, Wm. Guy Weigold, and Matthew J. Budoff

3

Radiation Dosimetry and CT Dose Reduction Techniques . . . . . . . . . . . . . . . . . . . 33 Kai H. Lee

4

Orientation and Approach to Cardiovascular Images . . . . . . . . . . . . . . . . . . . . . . 47 Jerold S. Shinbane and Antreas Hindoyan

Part II

Cardiac CT

5

Assessment of Cardiovascular Calcium: Interpretation, Prognostic Value, and Relationship to Lipids and Other Cardiovascular Risk Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Harvey S. Hecht

6

Natural History and Impact of Interventions on CAC . . . . . . . . . . . . . . . . . . . . . 121 Paolo Raggi

7

Methodology for CCTA Image Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 Mathew J. Budoff, Jerold S. Shinbane, and Songshao Mao

8

Post-processing and Reconstruction Techniques for the Coronary Arteries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Swaminatha V. Gurudevan

9

Coronary CT Angiography: Native Vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Stephan Achenbach

10

Coronary CT Angiography After Revascularization . . . . . . . . . . . . . . . . . . . . . . 179 Joachim Eckert, Marco Schmidt, Thomas Voigtländer, and Axel Schmermund

Part III

CT Angiography Assessment for Cardiac Pathology

11

Assessment of Cardiac Structure and Function by Computed Tomography Angiography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 John A. Rumberger

12

Cardiovascular CT for Perfusion and Delayed Contrast Enhancement Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Ravi K. Sharma, Ilan Gottlieb, and João A.C. Lima ix

x

Contents

13

Cardiovascular CT for Assessment of Pericardial/Myocardial Disease Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Muhammad Aamir Latif and Khurram Nasir

14

Computed Tomography Evaluation in Valvular Heart Disease . . . . . . . . . . . . . . 241 Nada Shaban, Javier Sanz, Leticia Fernández Friera, and Mario Jorge García

15

Transcatheter Aortic Valve Implantation (TAVI) . . . . . . . . . . . . . . . . . . . . . . . . . 255 Chesnal Dey Arepalli, Christopher Naoum, Philipp Blanke, and Jonathon A. Leipsic

16

Assessment of Cardiac and Thoracic Masses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Jabi E. Shriki, Patrick M. Colletti, and Suresh Maximin

Part IV

CT Vascular Angiography

17

CT Angiography of the Peripheral Arteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 Jabi E. Shriki, Leonardo C. Clavijo, and Gale L. Tang

18

Aortic, Renal, Mesenteric and Carotid CT Angiography . . . . . . . . . . . . . . . . . . . 319 Anas Alani and Matthew J. Budoff

19

Assessment of Pulmonary Vascular Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337 Bradley S. Messenger and Ronald J. Oudiz

Part V

Multidisciplinary Topics

20

Value Based Imaging for Coronary Artery Disease: Implications for Nuclear Cardiology and Cardiac CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349 Daniel S. Berman, Alan Rozanski, Piotr Slomka, Rine Nakanishi, Damini Dey, John D. Friedman, Sean W. Hayes, Louise E.J. Thomson, Reza Arsanjani, Rory Hachamovitch, James K. Min, Leslee J. Shaw, and Guido Germano

21

Coronary Computed Tomographic Angiography for Detection of Coronary Artery Disease: Analysis of Large-Scale Multicenter Trials and Registries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381 Leslee J. Shaw

22

Cardiothoracic Surgery Applications: Virtual CT Imaging Approaches to Procedural Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 Jerold S. Shinbane, Craig J. Baker, Mark J. Cunningham, and Vaughn A. Starnes

23

Computed Tomographic Angiography in the Assessment of Congenital Heart Disease and Coronary Artery Anomalies . . . . . . . . . . . . . . . . . . . . . . . . . . 429 Priya Pillutla and Stephen C. Cook

24

CCTA Cardiac Electrophysiology Applications: Substrate Identification, Virtual Procedural Planning, and Procedural Facilitation . . . . . . . . . . . . . . . . . . 455 Jerold S. Shinbane, Leslie A. Saxon, Rahul N. Doshi, Philip M. Chang, and Matthew J. Budoff

25

Cardiovascular CT: Interventional Cardiology Applications. . . . . . . . . . . . . . . . 487 Jeffrey M. Schussler

Contents

xi

26

Cardiovascular Magnetic Resonance Imaging: Overview of Clinical Applications in the Context of Cardiovascular CT . . . . . . . . . . . . . . . . . . . . . . . . 507 Jerold S. Shinbane, Jabi E. Shriki, Antreas Hindoyan, and Patrick M. Colletti

27

Cardiovascular CT in the Emergency Department . . . . . . . . . . . . . . . . . . . . . . . . 549 Asim Rizvi and James K. Min

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 561

Contributors

Stephan Achenbach, MD University of Erlangen, Erlangen, Germany Anas Alani, MD Department of Medicine, University of Florida – Gainesville, Gainsville, FL, USA Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA, USA Chesnal Dey Arepalli, MBBS, DNB Department of Radiology, St. Paul’s Hospital, Vancouver, BC, Canada Reza Arsanjani, MD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Craig J. Baker, MD, FACS Department of Surgery, Keck Hospital of the University of Southern California, Los Angeles, CA, USA Daniel S. Berman, MD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Philipp Blanke, MD Department of Medicine, St. Paul’s Hospital, Vancouver, BC, Canada Matthew J. Budoff, MD David Geffen School of Medicine at UCLA, Los Angeles Biomedical Research Institute, Torrance, CA, USA Phillip M. Chang, MD Department of Medicine, Keck Medical Center of USC/Keck School of Medicine at USC, Los Angeles, CA, USA Leonardo C. Clavijo, MD, PhD, FACC, FSCAI, FSVM Department of Medicine, Division of Cardiovascular Medicine, Department of Clinical Medicine, University of Southern California, Los Angeles, CA, USA Patrick M. Colletti, MD Department of Radiology, University of Southern California, Los Angeles, CA, USA Stephen C. Cook, MD, FACC Adult Congenital Heart Disease Center, Heart Institute, Children’s Hospital of Pittsburgh of UPMC, Pittsburgh, PA, USA Mark J. Cunningham, MD Department of Surgery, Keck Hospital of the University of Southern California, Los Angeles, CA, USA Damini Dey, PhD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Rahul N. Doshi, MD, FACC, FHRS Department of Medicine, Keck Medical Center of USC, Los Angeles, CA, USA Joachim Eckert, MD Department of Cardiology, Cardioangiologisches Centrum Bethanien, Frankfurt, Hessen, Germany

xiii

xiv

John D. Friedman, MD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Leticia Fernández Friera, MD Department of Medicine, Division of Cardiology, Mount Sinai Medical Center, New York, NY, USA Centro Nacional de Investigaciones Cardiovasculares, Madrid, Spain Mario Jorge Garcia, MD, FACC, FACP Division of Cardiology, Montefiore Medical Center, Bronx, NY, USA Guido Germano, PhD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Ilan Gottlieb, MD, MSc, PhD Casa de Saude Sao Jose, Rio de Janeiro, RJ, Brazil Swaminatha V. Gurudevan, MD, FACC Department of Medicine, Healthcare Partners Medical Group, Pasadena, CA, USA Rory Hachamovitch, MD Department of Nuclear Medicine, Cleveland Clinic, Heart and Vascular Institute, Cleveland, OH, USA Sean W. Hayes, MD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Harvey S. Hecht, MD, FACC, FSSCT Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, NY, USA Mount Sinai Medical Center, New York, NY, USA Antreas Hindoyan, MD Division of Cardiovascular Medicine, Department of Internal Medicine, Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA Muhammad Aamir Latif, MD Department of Medicine, Center for Healthcare Advancement and Outcomes, Baptist Health South Florida, Miami, FL, USA Kai H. Lee, PhD Associate Professor of Clinical Radiology, Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA Jonathon A. Leipsic, MD, FRCPC, FSCCT Department of Radiology, St. Paul’s Hospital, Vancouver, BC, Canada João A.C. Lima, MD Division of Cardiology, Johns Hopkins Hospital, Baltimore, MD, USA Songshou Mao, MD Department of Medicine, Los Angeles Biomedical Research Institute, Los Angeles, CA, USA Suresh Maximin, MD Department of Radiology, University of Washington, Seattle, WA, USA Bradley S. Messenger, MD Division of Cardiology, Department of Medicine, Harbor-UCLA Medical Center, Torrance, CA, USA James K. Min, MD, FACC Department of Radiology, Dalio Institute of Cardiovascular Imaging, Weill Cornell Medical College and the NewYork Presbyterian Hospital, New York, NY, USA Rine Nakanishi, MD, PhD Department of Medicine, Cardiac CT, Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA, USA Christopher Naoum, MBBS, FRACP Department of Radiology, St. Paul’s Hospital, Vancouver, BC, Canada

Contributors

Contributors

xv

Khurram Nasir, MD, MPH Department of Medicine, Center for Healthcare Advancement and Outcomes, Baptist Health South Florida, Miami Beach, FL, USA Ronald J. Oudiz, MD Department of Medicine, Los Angeles Biomedical Research Institute, The David Geffen School of Medicine at UCLA, Harbor-UCLA Medical Center, Torrance, CA, USA Priya Pillutla, MD Adult Congenital Heart Disease Program, Harbor-UCLA Medical Center, Torrance, CA, USA Paolo Raggi, MD Department of Medicine, Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, AB, Canada Asim Rizvi, MD Department of Radiology, Dalio Institute of Cardiovascular Imaging, Weill Cornell Medical College and the NewYork Presbyterian Hospital, New York, NY, USA Alan Rozanski, MD Division of Cardiology, Mt. Sinai Saint Luke’s and Roosevelt Hospitals, New York, NY, USA John A. Rumberger, PhD, MD Cardiac Imaging, The Princeton Longevity Center, Princeton, NJ, USA Javier Sanz, MD Department of Medicine, Division of Cardiology, Mount Sinai Medical Center, New York, NY, USA Leslie A. Saxon, MD Department of Medicine, USC Center for Body Computing, Keck Medical Center of USC, Los Angeles, CA, USA Axel Schmermund, MD Department of Cardiology, Cardioangiologisches Centrum Bethanien, Frankfurt, Hessen, Germany Marco J.M. Schmidt, MD Department of Cardiology, Cardioangiologisches Centrum Bethanien, Frankfurt, Hessen, Germany Jeffrey M. Schussler, MD, FACC, FSCAI, FSCCT, FACP Division of Cardiology, Department of Internal Medicine, Baylor University Medical Center, Dallas, TX/Jack and Jane Hamilton Heart and Vascular Hospital, Dallas, TX, USA Division of Cardiology, Department of Medicine, Texas A&M College of Medicine, Dallas, TX, USA Nada Shaban, MD Department of Medicine, Division of Cardiology, North Shore University Hospital, Manhasset, NY, USA Ravi K. Sharma, MD Division of Cardiology, Johns Hopkins Hospital, Baltimore, MD, USA Leslee Shaw, PhD Department of Medicine, Emory Clinical Cardiovascular Research Institute, Emory University School of Medicine, Atlanta, GA, USA Jerold S. Shinbane, MD, FACC, FHRS, FSCCT Division of Cardiovascular Medicine, Department of Internal Medicine, Keck School of Medicine of the University of Southern California, Los Angeles, CA, USA Jabi E. Shriki, MD Department of Radiology, Puget VA Health System, University of Washington, Seattle, WA, USA Piotr Slomka, PhD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Vaughn A. Starnes, MD H. Russell Smith Foundation, Cardiovascular Thoracic Institute, Keck Hospital of the University of Southern California, Los Angeles, CA, USA

xvi

Gale L. Tang, MD Department of Surgery, University of Washington, Seattle, WA, USA Louise E.J. Thomson, MBChB Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA Thomas Voigtländer, MD Department of Cardiology, Cardioangiologisches Centrum Bethanien, Frankfurt, Hessen, Germany Wm. Guy Weigold, MD, FACC, FSCCT Department of Medicine, Department of Medicine (Cardiology), Cardiac CT, MedStar Washington Hospital Center, Washington, DC, USA Cardiac CT Core Lab, MedStar Health Research Institute, Washington, DC, USA MedStar Cardiovascular Research Network, Washington, DC, USA

Contributors

Part I Overview

Computed Tomography

1

Matthew J. Budoff

Abstract

Cardiac CT scanners are rapidly improving, each major vendor has introduced a state of the art scanner every 2–3 years. The basic applications, terminology and acquisition has not changed dramatically, however, improvements in hardware and software continue to reduce radiation exposure, scan times, artifacts and improve image quality. This chapter outlines the basic CT terminology, functions and background behind the current state of CT scanners for cardiac applications. It reviews spatial, temporal and contrast resolution limits of the CT scanners. An overview of common terms, radiation exposure and protocols are included. This acts as an introductory chapter to be expanded by subsequent chapters that will each go into more details on specific topics. Comparison to magnetic resonance for image quality and functionality, and dose comparisons to mammography, nuclear and fluoroscopy are included. Keywords

Cardiac CT • Angiography • MDCT • MRI • Coronary calcium • Protocols • Radiation • Spatial resolution • Temporal resolution

Overview of X-ray Computed Tomography The development of computed tomography (CT), resulting in widespread clinical use of CT scanning by the early 1980s, was a major breakthrough in clinical diagnosis across multiple fields. The primary advantage of CT was the ability to obtain thin cross-sectional axial images, with improved spatial resolution over ultrasound, nuclear medicine, and magnetic resonance imaging. This imaging avoided superposition of three-dimensional (3-D) structures onto a planar 2-D representation, as is the problem with conventional projection X-ray (fluoroscopy). CT images, which are inherently digital and thus quite robust, are amenable to 3-D

M.J. Budoff, MD David Geffen School of Medicine at UCLA, Los Angeles Biomedical Research Institute, Torrance, CA, USA e-mail: [email protected]

computer reconstruction, allowing for ultimately nearly an infinite number of projections. From a cardiac perspective, the increased spatial resolution is the reason for its increase in sensitivity for atherosclerosis, plaque detection and coronary artery disease (CAD). With CT, smaller objects can be seen with better image quality. Localization of structures (in any plane) is more accurate and easier with tomography than with projection imaging like fluoroscopy. The exceptional contrast resolution of CT (ability to differentiate fat, air, tissue and water), allows visualization of more than the lumen or stent, but rather the plaque, artery wall and other cardiac and non-cardiac structures simultaneously. The basic principle of CT is that a fan-shaped, thin X-ray beam passes through the body at many angles to allow for cross-sectional imaging. The corresponding X-ray transmission measurements are collected by a detector array. Information entering the detector array and X-ray beam itself is collimated to produce thin sections while avoiding unnecessary photon scatter (to keep radiation exposure and

© Springer International Publishing 2016 M.J. Budoff, J.S. Shinbane (eds.), Cardiac CT Imaging: Diagnosis of Cardiovascular Disease, DOI 10.1007/978-3-319-28219-0_1

3

4

M.J. Budoff

Table 1.1 Typical Hounsfield unit values Air ~ −1000 HU Fat −100 to −40 Water – zero Non-enhanced myocardium and blood – 40–60 Contrast enhanced myocardium 80–140 Calcium >130 (to about 1000) Enhanced blood pools (lumen, aorta, LV) 300–500 Metal >1000

image noise to a minimum). The x-ray tub and detector array rotate around the patient separated by 180°, allowing continuous acquisition of data. The data recorded by the detectors are digitized into picture elements (pixels) with known dimensions. The gray-scale information contained in each individual pixel is reconstructed according to the attenuation of the X-ray beam along its path using a standardized technique termed “filtered back projection.” Grayscale values for pixels within the reconstructed tomogram are defined with reference to the value for water and are called “Hounsfield units” (HU; for the 1979 Nobel Prize winner, Sir Godfrey N. Hounsfield), or simply “CT numbers.” These CT numbers are the attenuation or brightness of the individual pixel (smallest definable unit on CT) of data. A three dimensional pixel is called a voxel. Typical pixel values for studies commonly seen on cardiac CT are listed in Table 1.1. Dr Hounsfield is credited with the invention of the CT scanner in late 1960s. Since CT uses X-ray absorption to create images, the differences in the image brightness at any point will depend on physical density and the presence of atoms with a high difference in anatomic number like calcium, and soft tissue and water. The absorption of the X-ray beam by different atoms will cause differences in CT brightness on the resulting image (contrast resolution). Blood and soft tissue (in the absence of vascular contrast enhancement) have similar density and consist of similar proportions of the same atoms (hydrogen, oxygen, carbon). Bone has an abundance of calcium and is thus brighter on CT. Fat has an abundance of hydrogen. Lung contains air which is of extremely low physical density and appears black on CT (HU −1000). The higher the density, the brighter the structure on CT. Calcium is bright white, air is black, and muscle or blood is gray. There are over 5000 shades of this gray scale represented on CT, centered around zero (water-gray). Computed tomography, therefore, can distinguish blood from air, fat and bone but not readily from muscle or other soft tissue. The densities of blood, myocardium, thrombus, and fibrous tissues are so similar in their CT number, that non-enhanced CT cannot distinguish these structures. Thus, the ventricles and other cardiac chambers can be seen on non-enhanced CT, but delineating the wall from the blood

Fig. 1.1 A non-contrast CT scan of the heart. Quite a bit of information can be garnered without contrast. The pericardium is visible as a thin line just below the R and L. The coronary arteries can be seen, and diameters and calcifications are present. The right coronary artery is seen near the R, the left anterior at the L, and the circumflex at the C. The four chambers of the heart are also seen, and relative sizes can be measured from this non-contrast study. The interatrial septum is clearly seen (red arrow). The ascending aorta is also present on this image and can be evaluated. Ao aorta, L left anterior descending artery, LA left atrium, LV left ventricle, RA right atrium, RV right ventricle

pool is not possible (Fig. 1.1). Investigators have validated the measurement of “LV size” with cardiac CT, which is the sum of both left ventricle (LV) mass and volume [1]. Due to the thin wall which does not contribute significantly to the total measured volume, the left and right atrial volumes can be accurately measured on non-contrast CT [2]. Because contrast resolution uses attenuation or density to visualize structures in gray scale, limitations of contrast resolution exist even on contrast enhanced studies. These include differentiating the cardiac vessels from cardiac cavities with same density (such as when the arteries run become intra-myocardial), and differentiating non-calcified plaque from surrounding low density structures, including thrombus. Even with good contrast enhancement, differentiating different types of plaque (lipid-laden and fibrous) can sometimes be challenging, although it is always easy to differentiate the bright white plaques (calcified) from non-calcific plaques.

1

Computed Tomography

5

Fig. 1.2 Sequential 3 mm slices from a non-contrast CT scan study (calcium scan). This study depicts the course and calcifications of the left anterior descending (LAD) artery. The white calcifications

are easily seen (red arrows) and quantitated by the computer to derive a calcium score, volume or mass with high inter-reader reproducibility

The higher spatial resolution of CT allows visualization of coronary arteries both with and without contrast enhancement. The ability to see the coronary arteries on a noncontrast study depends upon the fat surrounding the artery (of lower density, thus more black on images), providing a natural contrast between the myocardium and the epicardial artery (Fig. 1.1). Usually, the entire course of each coronary artery is visible on non-enhanced scans (Fig. 1.2). The major exception is bridging, when the coronary artery delves into the myocardium and cannot be distinguished without contrast. The distinction of blood and soft tissue (such as the left ventricle, where there is no air or fat to act as a natural contrast agent) requires injection of contrast with CT. Similarly, distinguishing the lumen and wall of the coronary artery also requires contrast enhancement. The accentuated absorption of X-rays by elements of high atomic number like calcium and iodine allows excellent visualization of small amounts of coronary calcium as well as the contrast-enhanced lumen of medium-size coronary arteries (Fig. 1.3). Air attenuates the X-ray less than water, and bone attenuates it more than water, so that in a given patient, Hounsfield units may range from −1000 HU (air) through 0 HU (water) to above +1000 HU

(bone cortex), Table 1.1. Coronary artery calcium in coronary atherosclerosis (consisting of the same calcium phosphate as in bone) has CT number >130 HU, typically going as high as +1000 HU. It does not go as high as the bony cortex of the spine due to the smaller quantity and mostly inhomogeneous distribution in the coronary artery plaque. Metal, such as that found in valves, wires, stents and surgical clips, typically have densities of +1000 HU or higher.

Cardiac CT Cardiac computed tomography (CT) provides image slices or tomograms of the heart. CT technology has significantly improved since its introduction into clinical practice in 1972. Current conventional scanners used for cardiac and cardiovascular imaging now employ either a rotating X-ray source with a circular, stationary detector array (spiral or helical CT) or a rotating electron beam (EBCT). The attenuation map recorded by the detectors is then transformed through a filtered back-projection into the CT image used for diagnosis. The biggest issue with cardiac imaging is the need for

6

M.J. Budoff

RCA LAD Collateral Ramus

Fig. 1.3 A contrast-enhanced CT of the coronary arteries, with excellent visualization of a high-grade stenosis in the mid-portion of the LAD. A large collateral vessel is seen from the RCA, but this is quite rare, as usually the collaterals are too small to be well seen on cardiac CT. A large ramus intermedius is well visualized, and the dominant RCA is present. This is but one view of many that can be visualized with cardiac CT, allowing for near-complete visualizations of the coronary tree

both spatial and temporal resolution. Cardiac magnetic resonance (MR) has been an emerging technique for almost two decades, making little progress toward widespread utilization over this time. Temporal resolution (how long it takes to obtain an image) is inversely related to spatial resolution with cardiac MR. Improving the MR spatial resolution requires prolonging the imaging time. This greatly limits the ability to focus with precision on moving objects, as the viewer needs to settle for either a high resolution image plagued by cardiac motion, or a low resolution image with no motion artifacts. Cardiac CT does not suffer from this inverse relationship, and allows for both high spatial and temporal resolution simultaneously. Multidetector CT (MDCT), with improved spatial resolution, allows for rotation speeds now on the order of 130–220 ms. The most distinct advantage of cardiac CT over cardiac MR is the improved spatial resolution and thinner slice thickness achievable with current systems. CT has the ability to image every 0.5 mm (submillimeter slices), providing high z-axis (through plane resolution). In-plane resolution is dependent upon the number of pixels that can be seen by a given detector array. Resolution of current CT systems uses a matrix of 512 × 512, allowing x- and y-axis (in plane) resolution down to 0.35 mm. MR systems use a matrix of 256 × 256, and flat plate technology currently used in advanced fluouroscopy

labs and cardiac catheterization labs use 1024 × 1024 matrix resolution. The best resolution reported by a cardiac MR study (using the 3 Tesla magnet) demonstrated resolution in the x-, y- and z-axes of 0.6 × 0.6 × 3 mm [3]. The best resolution offered by cardiac CT is 0.35 × 0.35 × 0.5 mm, which is almost a factor of 10 better spatial resolution and approaching the ultimate for 3-D tomography of nearly cubic (isotropic) “voxels” – or volume elements (a three dimensional pixel). As we consider non-invasive angiography with either CT or MR, we need to remember that both spatial and temporal resolution is much higher with traditional invasive angiography (discussed in more detail below). Reconstruction algorithms and multi-“head” detectors common to both current electron beam and spiral/helical CT have been implemented enabling volumetric imaging, and multiple high-quality reconstructions of various volumes of interest can be done either prospectively or retrospectively, depending on the method. The details of each type of scanner and principles of use will be described in detail.

MDCT Methods Sub-second MDCT scanners use a rapidly rotating X-ray tube and several rows of detectors, also rotating. The tube and detectors are fitted with slip rings that allow them to continuously move through multiple 360° rotations. The “helical” or “spiral” mode is possible secondary to the development of this slip-ring interconnect. This allows the X-ray tube and detectors to rotate continuously during image acquisition since no wires directly connect the rotating and stationary components of the system (i.e., no need to unwind the wires). This slip-ring technology was considered the most fundamental breakthrough for CT, allowing advancement from conventional CT performing single slice scanning in the 1980s to rapid multislice scanning in the 1990s. With the gantry continuously rotating, the table moves the patient through the imaging plane at a predetermined speed (table speed). The relative speed of the gantry relative to the rotation of the detectors is the scan pitch. Pitch is calculated as table speed divided by collimator width. The collimator width is simply the number of detectors multiplied by the detector width. A typical 64 detector system with 0.625 mm detectors will thus have a collimation width of 64 × 0.625 mm = 40 mm. Thus, each rotation of the detector array will ‘cover’ 40 mm of the body. If the pitch is 1, than the table is moving at 40 mm and the coverage is 40 mm, allowing for no overlapping data and acquisition of 0.625 mm data per slice. Moving the table faster will lead to wider slices, as a pitch of 1.5 will infer that the table is moving at 60 mm/rotation, while the detector array only covers 40 mm, so each of the 64 detectors will be responsible for almost 1 mm of the 60 mm that was covered during the

1

Computed Tomography

rotation. A low pitch (low table speed, typically used in cardiac imaging) allows for over-lapping data from adjacent detectors. A typical pitch for cardiac CT is 0.25, meaning that the table is moved at 10 mm per rotation, while the detectors cover 40 mm, allowing thin slice acquisition and overlapping datasets. The heart is literally moved only ¼ of the way through the detector array each rotation, so it takes 4 rotations to completely cover any portion of the heart. The pitch is varied based upon the heart rate of the patient, to allow optimal timing of image acquisition. Most commonly, physicians use a low table speed and thin imaging, leading to a lot of images, each very thin axial slices which are of great value for visualizing the heart with the highest resolution. The downside is that the slower the table movement (while still rotating the X-ray tube), the higher the radiation exposure (See Chap. 3). The smooth rapid table motion or pitch in helical scanning allows complete coverage of the cardiac anatomy in 5–25 s, depending on the actual number of multi-row detectors. The current generation of MDCT systems complete a 360° rotation in about 3 tenths of a second (300 ms) and are capable of acquiring 64–640 sections of the heart simultaneously with electrocardiographic (ECG) gating in either a prospective or retrospective mode. MDCT differs from single detector-row helical or spiral CT systems principally by the design of the detector arrays and data acquisition systems, which allow the detector arrays to be configured electronically to acquire multiple adjacent sections simultaneously. For example, in 64-row MDCT systems, 64 sections can be acquired at either 0.5–0.75 or 1–1.5 mm section widths or 16 sections 2.5 mm thick (commonly used for calcium scoring). In MDCT systems, like the preceding generation of single-detector-row helical scanners, the X-ray photons are generated within a specialized X-ray tube mounted on a rotating gantry. The patient is centered within the bore of the gantry such that the array of detectors is positioned to record incident photons after traversing the patient. Within the X-ray tube, a tungsten filament allows the tube current to be increased (mA), which proportionately increases the number of X-ray photons for producing an image. Thus, heavier patients can have increased mA, allowing for better tissue penetration and decreased image noise. One of the advantages of MDCT is the variability of the mA settings, thus increasing the versatility for general diagnostic CT in nearly all patients and nearly all body segments. The other variable on the acquisition is the voltage, commonly expressed as killivoltage (kv or kVp). The voltage was not varied on cardiac CT for the first 20 years of use, but more recently it has been noted that by reducing the kV, exponential reduction in radiation exposure can be achieved. As the kV goes down, image noise goes up, so it is important that the kV only be reduced on thinner patients. While 120 kV was most typical,

7

now increasingly 100 kV and even 80 kV studies are being reported, especially in children, where radiation issues are much more acute. For example, the calcium scanning protocol employed in the National Institutes of Health (NIH) Multi-Ethnic Study of Atherosclerosis is complex [4]. Scans were performed using prospective ECG gating at 50 % of the cardiac cycle, 120 kV, 106 mAs, 2.5 mm slice collimation, 0.5 s gantry rotation, and a partial scan reconstruction resulting in a temporal resolution of 300 ms. Images were reconstructed using the standard algorithm into a 35 cm display field of view. For participants weighing 100 kg (220 lb) or greater, the milliampere (mA) setting was increased by 25 %. However, the kV is typically not currently reduced for calcium scoring for two reasons. First, the radiation dose of calcium scoring is generally low (approximately 0.7 milliSieverts), similar to mammography (0.75 milliSieverts) and lower than annual background radiation (3–6 milliSieverts per year). Secondly, as the kV is lowered, calcium and contrast appear brighter, and this would change the calcium scores, which up until this point, have only been obtained using 120 kV acquisitions. As more data is available on the algorithms for scoring with lower kV scans, calcium scoring may undergo a radiation reduction of up to 40 % by lowering the kV from 120 to 100 for the acquisition. This will require new thresholds for definitions of calcium (for example- 147 HU instead of 130 HU as the definition of calcium, and 228 instead of 200 HU for the next threshold, etc.). The exact thresholds will have to be developed for each CT system prior to use. Thus, while calcium score dose can be reduced by iterative reconstruction, use of 100 kVp will need to wait for further validation. This is not an issue with CT angiography, as lower kVp will brighten the contrast, raising the signal to noise quality of the study. MDCT systems can operate in either the sequential (prospective triggered) or helical mode (retrospective gating). These modes of scanning are dependent upon whether the patient on the CT couch is stationary (axial, or sequential mode) or moved at a fixed speed relative to the gantry rotation (helical mode). The sequential mode utilizes prospective ECG triggering at predetermined offset from the ECGdetected R wave analogous to EBCT and is the current mode for measuring coronary calcium at most centers using MDCT, and increasingly being used for CT angiography when heart rates are stable and slow. This mode utilizes a “step and shoot modality,” which reduces radiation exposure by “prospectively” acquiring images, as compared to the helical mode, where continuous radiation is applied (and thousands of images created) and images are “retrospectively” aligned to the ECG tracing. In the sequential mode, a 64-slice scanner can acquire 64 simultaneous data channels of image information gated prospectively to the ECG. Thus a 64-channel system (with 0.625 mm detectors) can acquire,

8

M.J. Budoff

within the same cardiac cycle, 40 mm in coverage per heartbeat (collimation). The promise of improved cardiac imaging from the 64–640 slice scanners is mostly larger volumes of coverage simultaneously (up to 160 mm of coverage per rotation with a 320 slice scanner with 0.5 mm detectors), allowing for less z-axis alignment issues (cranial– caudal), and improved 3-D modeling with only 2–5 s of imaging, although each vendor has a different array of detectors, with different slice widths and capabilities (Table 1.2). As coverage speeds increase, breath-hold and contrast requirements will also diminish. Modern MDCT systems have currently an X-ray gantry rotation time of less than 500 ms. The fastest available rotation time is 260 ms, by using half scan reconstruction Table 1.2 Sample protocols for MDCT angiography: contrastenhanced retrospectively ECG-gated scan 4-Slice scanner: 4 × 1.0 mm collimation, table feed 1.5 mm/rotation, effective tube current 400 mAs at 120 kV. Pitch =1.5/4.0 collimation =0.375. Average scan time =35 s 16-Slice scanner (1.5 mm slices): 16 × 1.5 mm collimation, table feed 3.8 mm/rotation, effective tube current 133 mAs at 120 kV. Pitch =3.8/24 mm collimation =0.16. Average scan time =15–20 s 16-Slice scanner (0.75 mm slices): 16 × 0.75 mm collimation, table feed 3.4 mm/rotation, effective tube current 550–650 mAs at 120 kV. Pitch =3.4/12 mm collimation =0.28. Average scan time = 15–20 s 64-Slice scanner (0.625 mm slices): 64 × 0.6.25 mm collimation, table feed 10 mm/rotation, effective tube current 685 mAs at 120 kV. Pitch =10/40 mm collimation =0.25. Average scan time = 5 s Dual Source scanner (0.6 mm slices): 32 × 0.6 mm collimation, table feed 6 mm/rotation, effective tube current 685 mAs at 120 kV. Pitch =6/19.2 mm collimation =0.3. Average scan time = 10–12 s 320-Slice scanner (0. 5 mm slices): 320 × 0.5 mm collimation, table feed 12 mm/rotation, effective tube current 685 mAs at 120 kV. Pitch =12/40 mm collimation =0.3. Average scan time = 2–3 s

(discussed below), lowers this to 130 ms acquisitions. This remains suboptimal in faster heart rates (>70 bpm), as imaging during systole (or atrial contraction during late diastole) will be plagued by motion artifacts. Reconstruction algorithms have been developed that permit the use of data acquired during a limited part of the X-ray tube rotation (e.g., little more than one half of one rotation or smaller sections of several subsequent rotations) to reconstruct one cross-sectional image (described below). Simultaneous recording of the patient ECG permits the assignment of reconstructed images to certain time instants in the cardiac cycle. Image acquisition windows of approximately 200 ms can be achieved without the necessity to average data acquired over more than one heartbeat (Fig. 1.4). This may be sufficient to obtain images free of motion artifacts in many patients if the data reconstruction window is positioned in suitable phases of the cardiac cycle, and the patient has a sufficiently low heart rate. Motion-free segments on fourslice MDCT decrease from approximately 80 to 54 % with increasing heart rates [5], and similar observations have been made with both 16- and 64-detector systems. Dual source CT, utilizing two x-ray tubes and two detector arrays moving simultaneously around the body, can utilize partial images from each detector array to effectively ‘half’ the temporal resolution, allowing motion free images up to heart rates of 70 bpm or more. However, the system is limited by 32 detectors, so collimation or coverage is only 19.2 mm per rotation (32 × 0.6 mm).

MDCT Terminology Isotropic Data Acquisition The biggest advance that the newest systems provide is thinner slices, important for improving image quality as well as diminishing partial volume effects. The current systems

Detector 1 Detector 2

Fig. 1.4 A typical acquisition using the “halfscan” method on multidetector CT (Light-Speed16, GE Medical Systems, Milwaukee Wisconsin). This demonstrates an acquisition starting at approximately 50 % of the R-R interval. A scanner with a rotation speed of 200 ms takes approximately 250 ms to complete an image, as depicted

Detector 3 Detector 4

Segment: ~250 ms on LightSpeed16 /Ultra/Plus

1

Computed Tomography

allow for slice thick-nesses between 0.5 and 0.625 mm (depending on manufacturer and scan model). Thus, the imaging voxel is virtually equal in size in all dimensions (isotropic). The spatial resolution of current CT systems is 0.35 × 0.35 mm, and has always been limited by the z-axis (slice thickness). Current systems theoretically allow for isotropic resolution (as reconstructed images can be seen at 0.4 mm), allowing for no loss of data by reconstructing the data in a different plane. This is very important for imaging the coronary, peripheral, and carotid arteries, as they run perpendicular to the imaging plane (each slice only encompasses a small amount of data) and to follow these arteries, one must add multiple slices together in the z-axis. The old limitations of CT (better interpretation for structures that run within the plane it was imaged, i.e., parallel to the imaging plane) are no longer present. There is now no loss of data with reformatting the data with multiplanar reformation (MPR) or volume rendering (VR). This differs significantly from MR, which due to thick slice acquisition (still > 1 mm), does not allow free rotation of the resultant in-plane images. Thus, acquisition for CT is quite simple, obtaining axial slices through the area of interest, with the ability to reconstruct a three-dimensional image that can be free rotated. MR requires acquisition of data within the plane of interest, so if a short axis image of the left ventricle is required, it must be obtained in that plane, not reconstructed from the axial data. Furthermore, the thinner slice imaging allows for less partial volume artifacts (different densities overlying one another, causing a mixed picture of brightness on the resultant scan) and less streaking and shadowing, prevalent from dense calcifications and metal objects (such as bypass clips, pace-maker wires, and stents).

Pitch In the helical or spiral mode of operation, a 64-MDCT system can acquire 64 simultaneous data channels while there is continuous motion of the CT table. The relative motion of the rotating X-ray tube to the table speed is called the scan pitch and is particularly important for cardiac gating in the helical mode. Increased collimator coverage allows for decreasing the pitch, without losing spatial resolution. The definition of pitch for the multidetector systems is the distance the table travels per 360° rotation of the gantry, divided by the dimension of the exposed detector array (the collimation, which is the slice thickness times the number of imaging channels). For example, a 64-slice system, with 64 equal detectors each of 0.625 mm, gives a collimation of 40 mm. Thus, if the table is moving at 40 mm/rotation, the pitch is 1.0. The pitch remains 1.0 if the table moves at 60 mm/rotation and the slices are thicker (0.975 mm each), or if the number of channels (detectors utilized) increases. Moving the table faster will lead to thicker slices, which will decrease resolution and lead to partial volume effects. If

9

there is no overlap, the pitch is 1. If 50 % overlap is desired, the pitch is 0.5, as the table is moved slower to allow for overlapping images. This is necessary with multisector reconstruction. Typical pitch values for cardiac work are 0.25–0.4, allowing for up to a 4-fold overlap of images. If the collimation with 64-row scanners increases to 40 mm, the table can move four times faster than the 16-slice scanner, without affecting slice thickness. Thus, the coverage with increased numbers of detectors can go up dramatically over a short period of scanning time, by imaging more detectors and increasing the speed of table movement in concert.

Field of View Another method to improve image quality of the CT angiograph (CTA) is to keep the field of view small. The matrix for CT is 512 × 512, meaning that is the number of voxels in a given field of view. This is significantly better than current MR scanners, accounting for the improved spatial resolution. If the field of view is 15 cm, than each pixel is 0.3 mm. Increasing the field of view to 45 cm (typical for encompassing the entire chest) increases each pixel dimension to 0.9 mm, effectively reducing the spatial resolution of each data-point 3-fold.

Contrast Finally, the high scan speed allows substantial reduction in the amount of contrast material. The high speed of the scan allows one to decrease the amount of contrast administered; by using a 64-channel unit with a detector collimation of 0.625 mm and a tube rotation of 0.35 s (typical values for a 64-detector coronary CTA), the acquisition interval is around 5–6 s, which allows one to reduce the contrast load to approximately 50 mL. For a faster acquisition protocol, the contrast delivery strategy needs to be optimized according to the scan duration time. Use of a 320 detector scanner (which currently has 0.5 mm slices), covers the entire heart with one rotation, further reducing the contrast needs (although some minimal amount of contrast will be required to fill the heart and arteries in question). The general rule is the duration of scanning (scan acquisition time) equals the contrast infusion time. So, if an average rate of 4 mL/s is used, a 15-s scan acquisition would require 60 mL of contrast. With volume scanners (64and greater detector systems), the scan times are reduced to 5–8 s, and contrast doses are subsequently reduced as well.

Prospective Triggering The prospectively triggered image uses a “step and shoot” system, used for calcium scoring for years, now widely available for all MDCT scanners. This obtains images at a certain time of the cardiac cycle (see Chap. 2), which can be chosen in advance, and then only one image per detector per cardiac cycle is obtained. This reduces (radiation) requirements, and does allow for CT angiographic images

10

only when heart rates are slow, as motion artifacts may plague these images. When performing prospective gating, the temporal resolution of a helical or MDCT system is proportional to the gantry speed, which determines the time to complete one 360° rotation. To reconstruct each slice, data from a minimum of 180° plus the angle of the fan beam are required, typically 210° of the total 360° rotation. For a 64-row system with 0.35-s rotation, the temporal resolution is approximately 0.20 s or 200 ms. By reducing the display field of view to the 20 cm to encompass the heart, the number of views can be reduced to further improve temporal resolution to approximately 200 ms per slice. The majority of MDCT systems now have gantry rotation speeds of 250–330 ms and temporal resolution of 83–167 ms per image when used for measuring coronary calcium or creating individual images for CT angiography. Although physically faster rotational times may be possible in the future, this is still rotation of an X-ray source (with or without attached detectors) within a fixed radius of curvature. This is subject to the forces and limitations of momentum. While interpretation can be limited by the phase chosen (i.e.- 75 % of the R-R interval), there is an ability to add padding. Padding allows for additional phases to be imaged, but is dependent upon the heart rate. With slower heart rates, padding can be increased, to typically include an additional 10–15 % of the cardiac cycle on each side of the phase chosen (allowing for phases from 60 to 85 % to be acquired), so acquisition may be widened to allow extra phases to allow for some correction of motion artifacts. However, the more padding applied, the higher the radiation exposure.

Retrospective Gating The ECG is used to add R peak markers to the raw data set. A simultaneous ECG is recorded during the acquisition of cardiac images. The ECG is retrospectively used to assign source images to the respective phases of the cardiac cycle (ECG gating). The best imaging time to minimize coronary motion is from 40 to 80 % of the cardiac cycle (early to mid-diastole). The interval between markers determines the time of each scanned cardiac cycle. Retrospective, phase-specific, short time segments of several R-R intervals are combined to reconstruct a “frozen axial slice.” During helical scanning, the patient is moved through the CT scanner to cover a body volume (i.e., the heart). An advantage of the helical acquisition mode is that there is a continuous model of the volume of interest from base to apex, as opposed to the sequential/ cine mode in which there are discrete slabs of slices which have been obtained in a “step and shoot” prospective fashion. The obvious detriment to the helical acquisition is the increase in radiation dose delivered to the patient, as continuous images are created, and then “retrospectively”

M.J. Budoff

aligned to the ECG tracing to create images at any point of the R-R interval (cardiac cycle). This allows for all phases (1–100 %) to be recreated as needed, allowing for many phases to overcome motion artifacts, correct for stair-step (misalignment) artifacts, and calculate ejection fraction and wall motion. However, radiation doses are higher than prospective imaging, even when using dose modulation.

Halfscan Reconstruction A multislice helical CT halfscan (HS) reconstruction algorithm is most commonly employed for cardiac applications. Halfscan reconstruction using scan data from a 180° gantry rotation (180–250 ms) for generating one single axial image (Fig. 1.4). The imaging performances (in terms of the temporal resolution, z-axis resolution, image noise, and image artifacts) of the heartscan algorithm have demonstrated improvement over utilizing the entire rotation (full scan). It has been shown that the halfscan reconstruction results in improved image temporal resolution and is more immune to the inconsistent data problem induced by cardiac motions. The temporal resolution of multislice helical CT with the halfscan algorithm is approximately 60 % of the rotation speed of the scanner. The reason it is not 50 % of the rotation speed is that slightly more than 180° is required to create an image, as the fan beam width (usually 30°) must be excluded from the window (Fig. 1.4). Thus approximately 210° of a rotation is needed to reconstruct an entire image. Central time resolution (the point in the center of the image) is derived by the 180° rotation. MDCT using the standard halfscan reconstruction method permits reliable assessment of the main coronary branches (those in the center of the image field) in patients with heart rates below 65 beats/min [6,7]. The necessity of a low heart rate is a limitation of MDCT coronary angiography using this methodology. With halfscan reconstruction, the proportion of the acquisition time per heartbeat is linearly rising from 20 % at 60 beats/min to 33 % at 100 beats/min. When evaluating the diastolic time, the proportion is much greater. Slower heart rates have longer diastolic imaging. The diastolic time useful for imaging for a heart rate of 60 beats/min is on the order of 400 ms (excluding systole and atrial contraction). Thus, a 250 ms scan will take up 70 % of the diastolic imaging time. Increase the heart rate to 100 beats/min (systole remains relatively fixed) and the biggest change is shortening of diastole. Optimal diastolic imaging times are reduced to approximately 100 ms, clearly far too short for a motion-free MDCT scan acquisition of 250 ms (Fig. 1.5, left panel). Thus, heart rate reduction remains a central limitation for motion-free imaging of the heart using MDCT. This can be partially overcome by multisegment reconstruction, described below.

11

Computed Tomography

Fig. 1.5 LightSpeed16 CT angiography images. On the left is the halfscan reconstruction. On the right, is a reconstructed image using the same dataset, using multisegment reconstruction. There are still some

Detector 1 Detector 2

Multisegment Reconstruction The scan window refers to the time when the volume of interest is present in the scan field and is therefore “seen” by the detector. It is the limiting factor for temporal resolution. In multisegment reconstruction, it defines the number of cardiac cycles available and hence the maximum number of segments that can be used to reconstruct one slice. Multisegment reconstruction utilizes a helical scanning technique coupled with ECG synchronization (images are retrospectively aligned to the ECG data acquired to keep track of systolic and diastolic images). Multisegment

Det 4

4 sectors: ~65 ms Temporal resolution

Det 3

Detector 3 Detector 4 Det 2

Fig. 1.6 A demonstration of a theoretical image using multisegment reconstruction. The resulting image is constructed of four equal segments from four different detectors. Each is reconstructed from the same point in the cardiac cycle (approximately 50 % in this depiction). Four different detectors, each visualizing the same portion of the heart in the same portion of the cardiac cycle, can be used to add together to create one image. In practice, the segments are not always of equal length, and four images are not always available for reconstruction

motion artifacts in the distal right (green arrow), but much improved over the halfscan image, which is not interpretable (red arrow)

Det 1

1

reconstruction can typically use up 2–4 different segments correlated to the raw data. By using four heartbeats to create an image, the acquisition time can be reduced to a minimum of 65 ms (Fig. 1.6). During retrospective segmented reconstruction, views from different rotations are combined to simulate one halfscan rotation (approximately 210° of data is needed to create an image). To calculate the number of segments that can be extracted from a scan window, the number of positions available for reconstruction is determined automatically by a workstation. Each position is extended into a wedge so that the combination of all wedges simulates a halfscan tube

12

M.J. Budoff

Fig. 1.7 The figure on the left demonstrates a standard halfscan reconstruction with a rotation speed of 400 ms (approximate 260 ms image temporal resolution), with image data acquired on a four-channel MDCT system, 1.25 mm slice collimation, 0.6 gantry speed and a heart rate of 72 beats/min. The image to the right demonstrates the same heli-

cal scan data but processed with a two-sector reconstruction algorithm resulting in an effective temporal resolution of 180 ms. Note how the proximal right coronary artery (white arrows), as well as the left circumflex and great coronary vein, are now distinguishable (white arrowhead) and motion-free on the multisector reconstructed image

rotation. The raw data acquired in this virtual halfscan rotation is sufficient for the reconstruction of one slice. The size of the largest wedge (largest segment) defines the temporal resolution within the image (Fig. 1.6). In other words, the subsegment with the longest temporal data acquisition determines the temporal resolution of the overall image. If the four segments used to create an image were of the following size (65 ms, 65 ms, 50 ms, and 100 ms), then the temporal resolution of the reconstructed image is 100 ms. The use of multisegment reconstruction has allowed for markedly improved effective temporal resolution and image quality. Just to be clear about how this works, let’s use an analogy of a pie. Imagine needing just over half of a pie for a picture. To create an image, you can either add together one small slice from several pies (Fig. 1.6) or you can take one large piece from one pie (Fig. 1.4). The advantage of taking small pieces from each heartbeat is that the temporal resolution goes down proportionally. The difficulty in using this technique is that the pieces of the pie must align properly. Patients with even very slight arrhythmias (especially atrial fibrillation, sinus arrhythmia, or multifocal atrial rhythms), changing heart rates (increased vagal tone during breath-holding, catecholamine response after getting a contrast flush, etc.), or premature beats will cause misregistration (misalignment) to occur. If the heartbeats used are not perfectly regular, the computer will inadvertently add different portions of the cardiac cycle

together, making a non-diagnostic image. Thus, there is still need for regular rhythms with CT angiography, although with higher detector systems (i.e., 64–640 detectors), the number of heartbeats needed to cover the entire heart goes down to 1–4, reducing the chance of significant changes in heart rates due to premature beats, breath-holding, vagal or sympathetic tone. By combining information from each of the detector rows, the effective temporal resolution of the images can theoretically be reduced to as low as 65 ms. However, to do this requires four perfectly regular beats consecutively and a fast baseline heart rate, and usual reconstructions allow for two to three images to be utilized, yielding an effective temporal resolution of 130–180 ms per slice (Fig. 1.7), but with a direct proportional increase in radiation exposure to the patient. This method of segmenting the information of one image into several heartbeats is quite similar to the established prospective triggering techniques used for MRI of the coronary arteries [8,9]. With multisegment reconstruction the length of the acquisition time varies between 10 and 20 % of the R-R interval. Since the reconstruction algorithm is only capable of handling a limited number of segments, the pitch (table speed) is often increased for patients with higher heart rates. Thus, fewer images are available with higher heart rates, decreasing the potential success rate with this methodology (see section “Speed/temporal resolution” below).

1

Computed Tomography

Multisegment reconstruction has been shown to improve depiction of the coronary arteries as compared to halfscan reconstruction [10,11] (Fig. 1.5). This methodology will improve temporal resolution, but high heart rates will still increase the motion of the coronaries, increasing the likelihood of image blurring and non-diagnostic images (Fig. 1.5, right panel still demonstrating some blurring of the mid-distal right coronary artery with heart rates of 70–75 beats/min). It is fairly common for patients with low heart rates at rest to increase the heart rate significantly at the time of CT angiography. This can occur due to three factors: anxiety about the examination, the breath-hold, or the warmth of the contrast infusion to the patient. Thus, there is still a need for somewhat reduced heart rates during MDCT angiography with all current reconstruction systems. Studies examining the image quality of multisegment and halfscan reconstruction in CT with four [12] and eight [11] detector rows showed similar image quality in both phantoms and patients. However, Dewey et al. [13] demonstrated that the accuracy, sensitivity, specificity, and rate of non-assessable coronary branches were significantly better using multisegment reconstruction in a 16-slice MDCT scanner. The authors attributed the difference to the higher image quality and resulting longer vessel length free of motion artifacts with multisegment reconstruction. The obvious advantage of multisegment reconstruction is achieved by reducing the acquisition window per heartbeat to approximately 160 ms on average, particularly useful for diagnostic images of the right coronary artery and circumflex artery (the two arteries that suffer the most from motion artifacts) [14]. Therefore, MDCT in combination with multisegment reconstruction does not always require administration of beta blockers to reduce heart rate. This improvement simplifies the procedure and expands the group of patients who can be examined with non-invasive coronary angiography using MDCT. The potential is for even greater application with aligning these multiple segments together with 64+ detector scanners, further improving the diagnostic rate with MDCT angiography, and coverage for more widespread applications such as gating the entire aorta or triple-rule out imaging (imaging the entire chest to evaluate pulmonary embolism, aorta and coronaries simultaneously).

Limitations of Multisegment Reconstruction Heart rates above 65 beats/min demonstrate the biggest benefit of multisegment reconstructions. The benefit of multisegment reconstruction in low heart rates has not been demonstrated, and some experts recommend avoiding this in lower heart rates to minimize radiation exposure. A drawback of multisegment reconstruction is the effective radiation dose, which is estimated to be 30 % higher than necessary

13

for halfscan reconstruction, resulting from the lower pitch (slower table speed, increasing the time the X-ray beam is on) needed with multisegment reconstruction. The results of studies indicate that multi-segment reconstruction has superior diagnostic accuracy and image quality compared with halfscan reconstruction in patients with normal heart rates (Fig. 1.7). Thus, multisegment reconstruction holds promise to make the routine use of beta blockers to reduce the heart rate before CT coronary angiography less necessary as a routine. One further limitation is that certain heart rates cannot undergo multisegment reconstruction if the R-R interval (in milliseconds) is an even multiple of the scanner rotational speed. If the heart rate and the rotation of the scanner are synchronous, the same heart phase always corresponds to the same angle segment, and a partial-scan interval cannot be divided into smaller segments. Finally, if the heart rate is unexpectedly irregular (i.e., Premature Ventricular Contractions (PVCs) or stress reaction to the dye causing increased heart rate during imaging), multi-segment reconstruction will not be successful and the diagnostic image quality will have to rely on the halfscan reconstruction. Some scanners (Philips Medical Systems is the first to introduce such proprietary software) have intrinsic programs which improve the success rate with multisegment reconstruction at increased heart rates. How well these systems work and how often are clinical questions still being answered.

EBCT Methods While now discussion is included mostly for historical reasons, it is important to remember that the origins of cardiac imaging with CT lies firmly with the Electron beam computed tomography (GE Healthcare, Waukegan, WI). This is a tomographic-imaging device developed over 25 years ago specifically for cardiac imaging. At Harbor-UCLA, we started performing CT angiography of the coronary arteries in January 1995, literally 10 years earlier than MDCT systems were able to image the coronary tree with similar accuracy. To date, and specifically over the past decade, there has been a huge increase in diagnostic and prognostic data regarding coronary artery imaging. To this day, most of the prognostic and diagnostic work done with coronary artery calcium scanning was done with EBCT. In order to achieve rapid acquisition times useful for cardiac imaging, these fourth-generation CT scanners have been developed with a non-mechanical X-ray source. This allows for image acquisition on the order of 50–100 ms, and with prospective ECG triggering, the ability to “freeze” the heart. Electron beam scanners use a fixed X-ray source, which consists of a 210° arc ring of tungsten, activated by bombardment from a magnetically focused beam of electrons fired from an

14

Fig. 1.8 Depiction of the e-Speed electron beam computed tomography scanner. The electron source emits a beam, which is steered magnetically through the detection coil, then reflected to tungsten targets (A B C D), where a fan-shaped X-ray beam is created and, after passing through the area of interest, seen by the detectors

electron source “gun” behind the scanner ring. The patient is positioned inside the X-ray tube, obviating the need to move any part of the scanner during image acquisition (Fig. 1.8). EBCT is distinguished by its use of a scanning electron beam rather than the traditional X-ray tube and mechanical rotation device used in current “spiral” single and multiple detector scanners (requiring a physically moving X-ray source around the patient). EBCT requires only that the electron beam is swept across the tungsten targets to create a fan beam of X-ray, possible in as short as 50 ms per image (20 frames/s). The electron beam is emitted from the cathode, which is several feet superior to the patient’s head, and then passes through a magnetic coil, which bends the beam so that it will strike one of four tungsten anode targets. The magnetic coil also steers the beam through an arc of 210°. The X-rays are generated when the electron beam strikes the tungsten anode target, then passes through the patient in a fan-shaped X-ray and strikes the detector array positioned opposite the anodes. This stationary multisource/split-detector combination is coupled to a rotating electron beam and produces serial, contiguous, thin section tomographic scans in synchrony with the heart cycle. There have been four iterations for EBCT since it was introduced clinically in the early 1980s. Since its initial introduction in 1982, it has been known as “rapid CT,” “cine CT,” “Ultrafast CT©,” Electron Beam CT, and “Electron Beam Tomography©.” The overall imaging methods have remained unchanged, but there have been improvements in data storage, data manipulation and management, data display, and spatial resolution. The original C-100 scanner was replaced in 1993 by the C-150, which was replaced by the C-300 in 2000. The current EBCT scanner, the e-Speed (GE/Imatron) was introduced in 2003. The e-Speed is a multislice scanner and currently can perform a heart or

M.J. Budoff

vascular study in one-half the total examination time required by the C-150 and C-300 scanners. The e-Speed, in addition to the standard 50 and 100 ms scan modes common to all EBCT scanners, is capable of imaging speeds as fast as 33 ms. A major limitation of this modality currently is the slice width, which is limited to 1.5 mm. Current MDCT scanners can obtain images in 0.5–0.75 mm per slice. Three imaging protocols are used with the EBCT scanner. They provide the format to evaluate anatomy, cardiovascular function, and blood flow. The imaging protocol used to study cardiovascular anatomy is called the volume scanning mode and is similar to scanning protocols employed by conventional CT scanners. Single or dual scans are obtained and then the scanner couch is incremented a predetermined distance. For non-contrast studies, the table increment is usually the width of the scan slice, so that there is no overlap imaging of anatomy. For contrast studies, especially those to be reconstructed three-dimensionally, table incrementation is usually less than the slice width, giving overlap of information to improve spatial resolution (see Chaps. 3 and 4). For example, moving the table forward 1 mm, while taking a 1.5 mm slice, gives 33 % overlap per image. Using the older scanners, with slice thickness of 3 mm, moving the table only 1.5 mm gives overlap to improve spatial resolution. This scanning mode is utilized with and without contrast enhancement and provides high spatial resolution of cardiovascular anatomy. This technique provides high resolution axial images, and is ideal for evaluation of the aorta, coronary arteries, and congenital heart disease. A 3-D arteriogram reconstructed from tomographic images has the potential for more complete visualization of the coronary arteries (Figs. 1.3 and 1.9). The end-systolic images can be compared to the end-diastolic images, allowing for accurate measurement of regional wall thickening and motion, myocardial mass (utilizing the known specific gravity of cardiac tissue), global and regional ejection fraction [15]. Importantly, since blood is being injected into the venous system, simultaneous enhancement of the right and left ventricle allows for excellent visualization of all cardiac chambers simultaneously, and measure of both right and left ventricular function and structure [16]. The flow mode imaging protocol acquires a single image gated to the electrocardiogram at a predetermined point in the cardiac cycle (e.g., end-diastole (peak of the R-wave)). Images can be obtained for every cardiac cycle or multiples thereof. Scanning is initiated before the arrival of a contrast bolus at an area of interest (e.g., left ventricular (LV) myocardium) and is continued until the contrast has washed in and out of the area. Time density curves from the region of interest can be created for quantitative analysis of flow (Fig. 1.10). The filling of different chambers can be visualized sequentially, allowing for visualization of flow into and out of any area of interest. This was the original methodology

1

Computed Tomography

15

Fig. 1.9 A contrast-enhanced electron beam angiogram demonstrating long segments of the left anterior descending (arrow). Images such as seen here can be created which are much more similar to a conventional coronary angiogram, if desired. C circumflex artery

employed to assess for graft patency, prior to the ability to create 3-D images. It should be noted that early studies dating back to 1983 have demonstrated saphenous vein graft patency with this technique, achieving an accuracy of approximately 90 % as compared to invasive angiography [17]. This technique is still commonly employed to detect shunts (Chap. 23), as well as to deter-mine the length of time it takes for contrast to travel from the arm vein at the site of injection to the central aortic root (allowing for accurate image acquisition of the high resolution contrast-enhanced images, Chap. 7).

MDCT Scanners Spatial Resolution Spatial resolution compares the ability of the scanners to reproduce fine detail within an image, usually referred to as the high contrast spatial resolution. Spatial resolution is important in all three dimensions when measuring coronary plaque. Even if limited to the proximal coronary arteries, the left system courses obliquely within the x–y imaging plane, while the right coronary artery courses through the x–y imaging plane. Simply put, one axial image may demonstrate 5 or more centimeters of the left anterior descending, while most images will demonstrate only a cross-section of the right coronary artery. For more on cardiac anatomy with examples, see Chap. 4. The in-plane resolution of MDCT systems are among the best of any imaging modality, higher than echocardiography, nuclear imaging and magnetic resonance imaging. The resolution in z dimension is determined by the detector width in MDCT (this may be thought of as

the measurement of the slice width for individual images). This is a “voxel” or volume element and it has the potential to be nearly cubic using MDCT. Current coronary artery calcification (CAC) scanning protocols vary between manufacturers from 2.5 to 3.0 mm for MDCT. For CT angiography, MDCT obtains images with 0.5–0.625 mm per axial slice. Thus, MDCT has a significant advantage in terms of spatial resolution, and results in less partial volume averaging than all other modalities. Also the principles of resolution of say a 1 mm vessel require that the slice width be 1 mm or less. Partial volume averaging occurs when a small plaque has dramatically different CT numbers related to whether it is centered within one slice or divided between two adjacent slices. Thus, thinner slices will have less partial volume averaging. The visualization of smaller lesions is only possible with smaller slice widths (Fig. 1.11). Modern MDCT systems permit simultaneous data acquisition in 64–640 parallel cross-sections with 0.50–0.625 mm collimation each. The in-plane spatial resolution is now as high as 17 line pairs/cm with new high definition detector arrays. However, conventional CT angiography still has higher temporal and spatial resolution, allowing for better visualization of the smaller arteries and collateral vessels. Modern angiographic equipment has a resolution of 40 line pairs per centimeter with a six-inch field of view, the usual image magnification for coronary angiography [18]. Thus, invasive coronary angiography still has a 3–4-fold better resolution than current MDCT systems. It is likely to remain this way until the perfection of flat panel detectors for CT – which would be akin

16

M.J. Budoff

R

Search region of interest

250 200 150 100 50 0 1 0

1 2

1 4

1 6

1 8

2 0

2 2

2 4

2 6

2 8

3 0

3 2

3 4

3 6

3 8

4 0

CT VALUE VS . TIME ( S )

Fig. 1.10 A flow or timing study. This study images the same level over time. A region of interest (in this study, the circle is placed in the ascending aorta) defines the anatomy to be measured. The graph below measures the Hounsfield units (HU) of that region of interest on each subsequent image. Initially, there is no contrast enhancement, and the measures are of non-enhanced tissue, around 50–60 HU. Contrast starts to arrive on this study around 16 s, and peaks at 22 s. The bright white structure next to the ascending aorta is the superior vena cava, filled with unmixed contrast

to the current state of the art in conventional angiography devices. However, three dimensional reconstructions and better contrast resolution (ability to see different densities) of CT over fluoroscopy, helps improve diagnostic accuracy of MDCT. Generally, the higher X-ray flux (mAs = tube current × scan time) and greater number and efficiency of X-ray detectors available with MDCT devices leads to images with better signal-to-noise ratio and higher spatial resolution when compared to earlier scanners. Early detection of calcified plaque is dependent upon distinguishing the plaque from image noise. Newer MDCT systems (64 detector +) have reduced image noise compared to older systems (8–18 noise/HU versus 24 noise/HU). Image noise CT has been shown to have an association with body mass index which may result in falsely identifying noise as calcified plaque or overestimation of true plaque burden. Typical values for mAs for MDCT angiography is on the order of 300–400. While CT scanners have difficulties with the morbidly obese patient, MDCT can increase the mAs (and kV) to help with

Fig. 1.11 Spatial resolution as measured on computed tomography. A phantom is imaged, and the smaller line pairs per centimeter are evaluated. Somewhat similar to an eye chart, the thinnest lines clearly seen define the spatial resolution of the scanner

tissue penetration. Other approaches to overcome image noise in obese patients, beyond increasing mA and kV include reconstructing thicker slices (slice thickness is inversely proportional to image noise) and use of iterative reconstruction (discussed below).

Speed/Temporal Resolution Cardiac CT is dependent upon having a high temporal resolution to minimize coronary motion-related imaging artifacts. By coupling rapid image acquisition with ECG gating, images can be acquired in specific phases of the cardiac cycle. Studies have indicated that temporal resolutions of 19 ms are needed to suppress all pulmonary and cardiac motion [19]. Interestingly, temporal resolution needs to be faster to suppress motion of the pulmonary arteries than for cardiac imaging. The study by Ritchie et al. demonstrated the need for 19 ms imaging to suppress pulmonary motion (despite breathholding), while needing 35 ms imaging to fully suppress motion for cardiac structures. This is most likely due to the accordion motion of the pulmonary arteries, whereby the motion of the heart causes the surrounding pulmonary arteries to be pulled in and out with each beat. This has led to some physicians to use cardiac gating during pulmonary embolism studies to improve resolution down to fourth and fifth generation branches of the pulmonary system. Cardiac MR motion studies of the coronary arteries demonstrate that the rest period of the coronary artery (optimal diastolic imaging time) varies significantly between individ-

1

Computed Tomography

17

utilizing more detectors (i.e., 4- versus 8- versus 16- versus 64-detector/channel systems) reduces scan time (i.e., breathhold time) and section misregistration (misalignment), but has no effect on temporal resolution.

Radiation Dose

Fig. 1.12 A typical motion artifact seen with calcium scans on MDCT images. The limited reproducibility of this technique is due to the star artifacts seen on this image (RCA, white arrow). Prospective gating is done, with halfscan reconstruction techniques. In cases of faster heart rates, motion artifacts seen here are commonplace. To limit radiation to a reasonable level for this screening text, no overlap or retrospective images are obtained, so multisegment reconstruction is not possible

uals with a range of 66–333 ms for the left coronary artery and 66–200 ms for the right coronary artery [20] and that for mapping coronary flow, temporal resolution of 23 ms may be required for segments of the right coronary artery [21,22]. Current generation cardiac CT systems which create images for measuring calcified plaque at 135–220 ms (prospectively gated MDCT) cannot totally eliminate coronary artery motion in all individuals. Motion artifacts are especially prominent in the mid-right coronary artery, where the ballistic movement of the vessel may be as much as two to three times its diameter during the twisting and torsion of the heart during the cardiac cycle (Fig. 1.12). It should be noted that the motion of the coronary artery during the cardiac cycle is a 3-D event with translation, rotation, and accordion type movements. Thus portions of the coronary artery pass within and through adjacent tomographic planes during each R-R cycle. Blurring of plaques secondary to coronary motion increases in systems with slower acquisition speeds. The resulting artifacts tend to increase plaque area and decrease plaque density and thus alter the calcium measurements. The artifacts may make those segments of the CT angiogram non-diagnostic, a problem that plagued up to 70 % of the early four-slice MDCT system studies [5,12]. The image quality achieved with cardiac CT is determined not only by the 3-D spatial resolution but also by the temporal resolution. The spatial resolution is directly related to the scan slice thickness and the reconstruction matrix. The temporal resolution, which determines the degree of motion suppression, is dependent on the pitch factor (which is determined by the table speed), the gantry rotation time, and the patient’s heart rate during the examination. As stated above,

Computed tomography utilizes X-rays, a form of ionizing radiation, to produce the information required for generating CT images. Although ionizing radiation from natural sources is part of our daily existence (background radiation including air travel, ground sources, and television), a role of healthcare professionals involved in medical imaging is to understand potential risks of a test and balance those against the potential benefits. This is particularly true for diagnostic tests that will be applied to healthy individuals as part of a disease screening or risk stratification program. In order for healthcare professionals to effectively advise individuals they must have an understanding of the exposure involved. The use of radiological investigations is an accepted part of medical practice, justified in terms of clear clinical benefits to the patient which should far outweigh the small radiation risks. Diagnostic medical exposures, being the major source of man-made radiation exposure of the population, add up to 50 % of the radiation exposure to the population. However, even small radiation doses are not entirely without risk. A small fraction of the genetic mutations and malignant diseases occurring in the population can be attributed to natural background radiation. The concept of “effective dose” was introduced in 1975 to provide a mechanism for assessing the radiation detriment from partial body irradiations in terms of data derived from whole-body irradiations. The effective dose for a radiological investigation is the weighted sum of the doses to a number of body tissues, where the weighting factor for each tissue depends upon its relative sensitivity to radiation-induced cancer or severe hereditary effects. It thus provides a single dose estimate related to the total radiation risk, no matter how the radiation dose is distributed around the body. Adoption of the effective dose as a standard measure of dose allows comparability across the spectrum of medical and non-medical exposures. “The effective dose is, by definition, an estimate of the uniform, whole-body equivalent dose that would produce the same level of risk for adverse effects that results from the non-uniform partial body irradiation. The unit for the effective dose is the sievert (Sv)” (www.fda.gov/cdrh/ct/rqu. html). Although it has many limitations, the effective dose is often used to compare the dose from a CT examination, a fluoroscopic examination, and the background radiation one experiences in a year. Units are either millirem (mrem) or millisievert (mSv); 100 mrem equals 1 mSv. The estimated dose from chest X-ray is 0.04 mSv, and the average annual

M.J. Budoff

18 Table 1.3 Common tests with estimated radiation exposures Test 1 Stress MIBI 1 LC spine 1 Barium enema 1 Upper GI 1 Abdominal X-ray 1 Dental X-ray 1 Cardiac catheterization

Calcium scan CT angiography CTA dose modulation Lung CT Abdomen/pelvis Body scan Virtual colon

Radiation dose (mSv) 6 1.3 7 3 1 0.7 2.5–10 Radiation dose Radiation dose for MDCT (mSv) for EBCT (mSv) 1–1.5 0.6 8–13 1–1.5 5–8 – 8 1.5 10 2 12 2.6 8–14 2–3

back-ground radiation in the United States is about 300 mrem or 3 mSv [23]. Table 1.3 shows the estimated radiation doses of some commonly used tests. The Food and Drug Administration (FDA) in describing the radiation risks from CT screening in general used the following language (www.fda.gov/cdrh/ct/risks.html): In the field of radiation protection, it is commonly assumed that the risk for adverse health effects from cancer is proportional to the amount of radiation dose absorbed and the amount of dose depends on the type of X-ray examination. A CT examination with an effective dose of 10 millisieverts (abbreviated mSv; 1 mSv =1 mGy in the case of x rays) may be associated with an increase in the possibility of fatal cancer of approximately 1 chance in 2000. This increase in the possibility of a fatal cancer from radiation can be compared to the natural incidence of fatal cancer in the US population, about 1 chance in 5. In other words, for any one person the risk of radiation-induced cancer is much smaller than the natural risk of cancer. Nevertheless, this small increase in radiation-associated cancer risk for an individual can become a public health concern if large numbers of the population undergo increased numbers of CT procedures for screening purposes. Since CT is the most important source of ionizing radiation for the general population, dose reduction and avoidance is of the utmost importance, especially for the asymptomatic person undergoing risk stratification, rather than diagnostic workup. Already, 50 % of a person’s lifetime radiation exposure is due to medical testing, and this is expected to go up with increased exposure to nuclear tests and CT scanning. Used as a tool in preventive cardiology, cardiac CT is increasingly being performed in the population of asymptomatic persons, a priori healthier individuals, where use of excessive radiation is of special concern.

The other variable involved in dose is the protocol used. The lowest radiation and most commonly applied is acquired by prospective triggering; that is, the X-rays are only on during the acquisition of data that will be used for the image. MDCT, however, can use prospective triggering or retrospective gating, 120 or 140 kilovolts (kV) (with protocols possible with lower kV depending upon future research and tube current improvements), and a wide range of mA. Retrospective gating means that the X-rays are on throughout the heart cycle. One drawback of MDCT is the potential for higher radiation exposure to the patient, depending on the tube current selected for the examination. The X-ray photon flux expressed by the product of X-ray tube current and exposure time (mAs). For example, 200 mA with 0.5 s exposure time yields 100 mAs in MDCT. In millisieverts, calcium scanning leads to an approximate dose of 0.9 mSv for MDCT (similar to the dose of another screening test, mammography – 0.7 mSv). This is in comparison to a conventional coronary angiogram, with mean doses of 8 mSv [23]. There are primary three factors that go into radiation dosimetry. The X-ray energy (kV), tube current (mA), and exposure time. The distance of the X-ray source from the patient is a fourth source, but, unlike fluoroscopy, this distance is fixed in all current CT scanners. Since MDCT angiography at the time utilized retrospective imaging, and since radiation is continuously applied while only a fraction of the acquired data is utilized, high radiation doses from retrospective CT studies (doses of 6–10 mSv/study) still play a role in decisions to utilize this modality in heart failure and atrial fibrillation, where retrospective studies are largely required [24–26]. The American Heart Association wrote a scientific statement for radiation doses [23] from cardiac studies and cited the following doses: retrospective CTA with dose modulation – 9 mSv; prospective triggered CTA – 3 mSv; stress-rest Sestamibi – 9 mSv; FDG PET scan 14 mSv and invasive coronary angiogram – 7 mSv [23]. Doses have come down dramatically with MDCT due to multiple approaches, including lower kVp (i.e. 100), dose modulation, and prospective triggering, now averaging 2.1 mSev for MDCT angiography in clinical practice [24]. Since the publication by Choi et al. [24], new detectors, iterative reconstruction and more aggressive use of ‘fast-pitch’ imaging has further reduced doses. It is conceivable that mean doses will dip below 1 mSv once use of these new techniques becomes widespread. Theoretically, since narrow collimation (beam widths) causes “overbeaming,” the dose efficiency is lower with four-slice scanners than with more detectors. Estimated efficiency goes from 67 % with 4 slices (1.25 mm, 5 mm coverage) to 97 % with 16 slices (at 1.25 mm, covering 20 mm). However, obtaining thinner slices will offset this gain, as obtaining more images will lead to higher radiation

1

Computed Tomography

doses. Typical studies with four-slice MDCT scanners obtained 600 images, while 16-slice scanners typically produce about 1200 scans, and 64-detector scanners are reporting over 2000 images produced per study. Newer MDCT studies report continue to show dramatic radiation reductions from the early 4-slice MDCT scanners and even from the more recent 64-MDCT studies [27]. The energy used can be altered in MDCT studies, and while typical imaging is done in the range of 120–140 kV, this can be varied with MDCT. The attenuation (densities) of calcium and iodine contrast agents increases with reduced X-ray energy (reduced kV settings). This can reduce the X-ray exposure, as the X-ray power emitted by the tube decreases considerably with reduced kV settings. To compensate for the lower power (and resultant increased noise of the image), the tube current (mA) can be increased. Scans with less than 120 kV tube voltage (i.e., 80 kV) can potentially maintain contrast-to-noise ratios that have been established for coronary calcium and CT angiography images, and significant lower radiation expo-sure. Preliminary experiments have demonstrated a reduction in radiation exposures of up to 50 % with use of 80 kV. Jakobs et al. demonstrated that the radiation dose for CAC scoring with use of 80 kV may be as low as 0.6 mSv [28]. As the noise will go up with 80 kV imaging, this may be too low for CT angiography in larger patients, but for children and slim adults, this affords a sub-millisievert dose for the first time with CTA. CT angiography protocols are much more common with 100 kV protocols, providing a reduction in radiation of 40 %, as the dose reduction from kV reduction is exponential, so a 20 % kV reduction affords a 40 % dose reduction. A very low kV (i.e.- 80) is most often used for pediatric patients. Similarly, for obese patients, where penetration is important, a kV of 135 or 140 can be utilized, but the dose will go up exponentially, so this is not widely employed. The power (mAs) used is directly proportional to the radiation dose. Higher mAs results in lower image noise, following the relationship: noise is inversely proportional to the square root of the mAs. Thus, limiting mAs can result in lower radiation, but higher noise. It is important to remember that reducing mAs too much to save radiation will increase noise to the point where the scan is non-diagnostic. Most centers use either an “automatic” mAs system, which adjusts the mAs based upon the image quality, or leave the mAs relatively fixed. Prospective triggering or “sequential” mode is typically used and strongly recommended for coronary calcium assessment with MDCT, due to the lower radiation doses to which the patient is exposed. However, the drawback of using MDCT prospective triggering is the inability to perform overlapping images, and longer image acquisition times. Thus, all CT vendors currently recommend retrospective image acquisition in the “helical” mode for performance

19

of MDCT angiography, despite the higher radiation doses. This is due to the requirement during retrospective imaging for slower table movement to allow for oversampling, for gap-less and motion-reduced imaging, as well as the possibility of multisegment reconstruction. Retrospective imaging allows the clinician to have multiple phases of the cardiac cycle available for reconstruction, to find the portion of the study with the least coronary motion. Many clinicians utilize multiple phases for reconstruction, with different phases used for different coronary arteries. However, constant irradiation is redundant, as most images are reconstructed during the diastolic phases of the heart cycle. Thus, a new method for reducing radiation dose with MDCT is to implement tube current modulation. Tube current is reduced during systole, when images are not utilized for reconstruction of images for MDCT angiography, by 80 %, and then full dose is utilized during diastole. Depending upon the heart rate, this may reduce radiation exposure by as much as 47 % [28], but with slower heart rates, this reduction will be less, as systole encompasses a shorter and shorter fraction of the cardiac cycle. Dose modulation protocols reduce radiation doses with MDCT by attempting to decrease beam current during systole, when images are not used for interpretation, and should be employed whenever possible [29]. Most of these protocols turn down the beam current (mA) during systole, so that diagnostic images are still available from roughly 40–80 % of the R-R interval (cardiac cycle). Dose modulation widely, but not universally employed, as it is harder to use with very fast heart rates, as the time to ramp up and ramp down the radiation dose becomes significant, and the ever shortening diastole becomes a smaller target. Also, irregular heart rates (frequent premature beats or atrial fibrillation) makes anticipation of the next R-R interval challenging, so most often dose modulation is turned off in this setting. The routine use of beta blockers makes even more sense in this setting, given the increased ability to use dose modulation more effectively and frequently with lower heart rates. A misconception is the potential loss of wall motion and other functional data with dose modulation. If dose modulation is used, there are systolic images, and while they are somewhat noisier (harder to read plaque), the image quality is still excellent for wall motion and ejection fraction assessment. An image from end-systole (early diastole) and end-diastole can be compared to calculated wall thickening, ejection fraction, and cardiac motion. Another method to decrease radiation exposure is to increase the pitch. The pitch is usually very low, on the order of 0.20–0.25 for CT angiography cases, and usually 1 for coronary calcium scanning. This low pitch raises radiation exposure, but is partially compensated by the more efficient dose of using the larger collimation available with increased

20

detectors. For individuals with different heart rates, different pitches can be used to obtain similar datasets. Thus, for those with slower heart rates, a faster pitch may allow for some reduction in cardiac exposure during these studies, and thus proportionally less radiation with dose modulation than for those with higher heart rates. The newest scanners available in 2014 now allow for the first time, acquisition of retrospective imaging with ‘fast-pitch’ acquisitions, moving the gantry very fast during the cardiac cycle and acquiring multiple datasets with a much lower radiation profile and improved resolution. This likely represents the next great breakthrough in cardiac CT imaging, allowing for full datasets to calculate ejection fraction, wall motion and plaque with radiation doses of 65 bpm), optimal image quality requires a 2- or 3-beat acquisition (and use of multi-cycle recon) which increases radiation exposure to approximately 13 (2-beat) to 19 (3-beat) mSv [11, 12]. Acquiring with a wide exposure window (“padding”) can ensure the capture of motion-free data, but increases radiation exposure (Fig. 2.6). A narrower exposure window can be used with preservation of image quality if heart rate is controlled and reduces radiation exposure.

High Pitch Helical Scanning Radiation exposure is inversely proportional to pitch in ECG-gated helical CT. Thus, increasing the pitch could dramatically lower radiation requirements in cardiac CT. In 2009, Siemens introduced an innovative scanning method using a high pitch helical mode which takes advantage of the dual-source design [13, 14] (Fig. 2.7). The high pitch (3.2–3.4) would normally produce gaps in the attenuation data using a single-source system, but, in a dualsource system, these gaps are compensated for by gathering data from the second detector. Scan time is less than one second, with radiation exposures now reported less than 1 mSv. Initial studies performed on the first-generation dual-source system proved the feasibility of the method, though the authors noted that these first-generation systems are not suitable for this high pitch technique. High pitch scanning with their new scanner, with faster gantry rotation (250 ms) and larger coverage (96 rows), has been reported to produce very good coronary image quality with radiation exposure of 200. In the 100. Rumberger et al. [35] demonstrated that higher calcium scores are associated with a greater specificity for obstructive disease at the expense of sensitivity; for example, a threshold score of 368 was 95 % specific for the presence of obstructive CAD. In 1764 persons undergoing angiography, the sensitivity and negative predictive value in men and women were >99 % [36]; a score of 0 virtually excluded patients with obstructive CAD. In a separate study of 1851 patients undergoing CAC scanning and angiography [37], CAC scanning by EBT in conjunction with pretest probability of disease derived by a combination of age, gender, and risk factors, facilitated prediction of the severity and extent of angiographically significant CAD in symptomatic patients. In a recent meta-analysis of 10,355 symptomatic patients who underwent cardiac catheterization and CAC, 0 CAC was noted in 1941. Significant obstructive disease, defined as >50 % diameter stenosis, was noted in 5805 (56 %). For CAC >0 and the presence of >50 % diameter stenosis, the following were reported: sensitivity 98 %, specificity 40 %, positive predictive value 68 %, and negative predictive value 93 % [38].

Prognostic Studies in Symptomatic Patients The prognostic value of extensive CAC (>1000) in symptomatic males with established advanced CAD was demonstrated in a 5-year follow-up study of 150 patients [39]. More recently, in a meta-analysis of 3924 symptomatic patients with a 3.5 year follow up, the cardiac event rate was

5

Assessment of Cardiovascular Calcium

2.6 %/year in those with CAC >0 and 0.5 %/year in 0 CAC patients [38]. However, in this era of coronary computed tomographic angiography (CCTA), CAC alone is not justified in the symptomatic population; CCTA will identify the noncalcified plaque and obstructive disease that may be noted in these patients, even with 0 CAC.

Clarification Despite the apparently reasonable specificities, which are similar to those of stress testing, it must be understood that the purpose of CAC scanning is not to detect obstructive disease and, therefore, it is inappropriate to even use “specificity” in the context of obstruction. Rather, its purpose is to detect subclinical atherosclerosis in its early stages, for which it is virtually 100 % specific.

105

smoking, respectively. In women, only CAC was linked to events, with a relative risk of 2.6; risk factors were not related. The presence of CAC provided prognostic information incremental to age and other risk factors. Shaw et al. [44] retrospectively analyzed 10,377 asymptomatic patients with a 5-year follow-up after an initial EBT evaluation. All-cause mortality increased proportional to CAC, which was an independent predictor of risk after adjusting for all of the Framingham risk factors (p < 0.001). Superiority of CAC to conventional Framingham risk factor assessment was demonstrated by a significantly greater area under the ROC curves (0.73 versus 0.67, p < 0.001). Greenland et al. [45] analyzed a population-based study of 1461 prospectively followed, asymptomatic subjects who

Key Prognostic Studies in Primary Prevention and Comparisons with Standard Risk Factor Paradigms The utility of CAC for risk evaluation in the asymptomatic primary prevention population is dependent on prognostic studies documenting the relative risk conferred by calcified plaque quantitation compared to conventional risk factors. Raggi et al. [40] demonstrated, in 632 asymptomatic patients followed for 32 months, an annualized event rate of 0.1 %/ year in patients with 0 scores, compared to 2.1 %/year with scores of 1–99, 4.1 %/year with scores of 100–400, and 4.8 %/year with scores >400. Thus, the annualized event rates associated with coronary calcium were in the range considered to warrant secondary prevention classification by the Framingham Risk Score (Fig. 5.4). The odds ratio conferred by a calcium percentile >75 % was 21.5 times greater than for the lowest 25 %, compared to an odds ratio of 7 for the highest versus lowest quartiles of National Cholesterol Education Program (NCEP) risk factors (Fig. 5.5). Wong et al. [41], in 926 asymptomatic patients followed for 3.3 years, noted a relative risk of 8 for scores >270, after adjusting for age, gender, hypertension, high cholesterol, smoking, and diabetes. Arad et al. [42], in 1132 subjects followed for 3.6 years, reported odds ratios of 14.3–20.2 for scores ranging from >80 to >600; these were 3–7 times greater than for the NCEP risk factors. In a retrospective analysis of 5635 asymptomatic, predominantly low to moderate risk, largely middle-aged patients followed for 37 ± 12 months, Kondos et al. [43] found that the presence of any CAC by EBT was associated with a relative risk for events of 10.5, compared to 1.98 and 1.4 for diabetes and

4.8 4.1 4

2.1 2

0.11 0 0

1-90

100-400

>400

Calcium scores

Fig. 5.4 Relationship of coronary artery calcium score to annual hard cardiac event rates in 632 asymptomatic patients undergoing EBT calcified plaque imaging. The solid line indicates the 2 %/year event rate consistent with secondary prevention risk

25 21.5 20 Odds ratio

CAC in Asymptomatic Patients

Percent/year cardiac events

6

EBT All NCEP risk factors

15

10 7

6.2 5

3.1 1

1

3.1

1

0 1st

2nd 3rd Calcium percentile quartiles

4th

Fig. 5.5 Odds ratios of coronary artery calcium and NCEP risk factor quartiles for annual hard cardiac event rates in asymptomatic patients undergoing coronary artery calcium imaging

H.S. Hecht

106 3.3

3.5 3

2.4

CAC 0 1–100 101–300 >300 Doubling

2.5 2

2 1.5 1

0.7

0.5

Table 5.2 Risk of coronary events associated with increasing coronary artery calcium after adjusting for standard risk factors in MESA

0.12

Annual rate 0.11 % 0.59 % 1.43 % 2.87 %

Events/no at risk 15/3409 39/1728 41/752 67/833

HR 1.0 3.61 7.73 9.67 1.26

P 1000 who were followed for 17 months [47] during which 35 patients (36 %) suffered a hard cardiac event (myocardial infarction or cardiac death). The annualized event rate of 25 % refuted the erroneous concept that extensive calcified plaque may confer protection against plaque rupture and events. In a younger cohort of 2000 asymptomatic Army personnel, Taylor et al. [48] demonstrated the powerful predictive value of CAC. There was a relative risk of 11.8 in patients with CAC >44 compared to those with 0 CAC, after correcting for the Framingham Risk Score. In a much more elderly

population (71 years), Vliegenthart et al. found a hazard ratio of 4.6 for CAC 400–1000 compared to 400 was associated with a hazard ratio of 9.2. In the largest study using coronary calcium percentile rather than absolute scores, Becker et al. [51] in 1724 patients followed prospectively for 3.4 years, reported hazard ratios for CAC percentile >75 % versus 0 % of 6.8 for men and 7.9 for women. The area under the ROC curve for CAC percentile (0.81) was significantly superior to the Framingham (0.66), European Society of Cardiology (0.65), and PROCAM risk scores (0.63). Eighty two percent of patients who developed myocardial infarction or cardiac death were correctly classified as high risk by CAC percentile, compared to only 30 % by Framingham, 36 % by the European Society of Cardiology, and 32 % by PROCAM. Perhaps the most important study is the Multiethnic Study of Atherosclerosis, an NHLBI sponsored prospective evaluation of 6814 patients followed for 3.8 years [23]. Compared to patients with 0 CAC, the hazard ratios for a coronary event were 7.73 for those with CAC 101–300, and 9.67 among participants for CAC >300 (P < 0.001) (Table 5.2; Fig. 5.7).

5

Assessment of Cardiovascular Calcium

107

Table 5.3 Characteristics and risk ratio for follow-up studies using coronary artery calcium in asymptomatic persons Author Arad [42] Park [108] Raggi [40] Wong [41] Kondos [43] Greenland [45] Shaw [44] Arad [46] Taylor [48] Vliegenthart [49] Budoff [50] Lagoski [53] Becker [51]

N 1173 967 632 926 5635 1312 10,377 5585 2000 1795 25,503 3601 1726

Mean age (years) 53 67 52 54 51 66 53 59 40–50 71 56 45–84 57.7

Follow-up duration (years) 3.6 6.4 2.7 3.3 3.1 7.0 5 4.3 3.0 3.3 6.8 3.75 3.4

Calcium score cutoff CAC >160 CAC >142.1 Top quartile Top quartile (>270) CAC CAC >300 CAC ≥400 CAC ≥100 CAC >44 CAC >1000 CAC >400 CAC >0 CAC >400

Comparator group for RR calculation CAC 10, compared to 1.92 for CAC 1–10, and 0.87 for 0 CAC. Finally, in a meta-analysis of 64,873 patients followed for 4.2 years, the coronary event rate was 1 %/year for the 42,283 with CAC >0, compared to 0.13 %/year in the 25,903 patients with 0 CAC [38]. Finally, in the Heinz Nixdorf Recall Study [55], 4487 subjects without CAD were followed for 5 years. Low ATP III risk was noted in 51.5 %, while 28.8 % and 19.7 % were at intermediate and high risk, respectively. The prevalence of low (75th vs ≤25th percentile was 11.1 (p < 0.0001) for men and 3.2 (p = 0.006) for women. Adding CAC to the ATP III categories improved

Table 5.4 Summary of CAC absolute event rates from 14,856 patients in five prospective studies CAC 0 1–100 100–400 >400 >1000

FRS risk Very low Low Intermediate High Very high

Ten years event rate 1.1–1.7 % 2.3–5.9 % 12.8–16.4 % 22.5–28.6 % 37 %

Abbreviations: CAC coronary artery calcium, FRS Framingham risk score

the AUC from 0.602 to 0.727 in men and from 0.660 to 0.723 in women, and led to a reclassification of 77.1 % of intermediate risk individuals (62.9 % into low risk, and 14.1 % into high risk group). The relative risk associated with doubling of the CAC score was 1.32 (95 % CI: 1.20– 1.45, p < 0.001) in men and 1.25 (95 % CI: 1.11–1.42, p < 0.0001) in women. In all of these studies, receiver operator characteristic curves for CAC were superior to the Framingham Risk Score and the annual event rate for CAC >100–400 exceeded the coronary artery disease equivalent of >2 %/year. Table 5.3 summarizes the relative risk results of the largest published outcome studies. Amalgamation of data from five large prospective randomized studies [23, 46, 49, 51, 55] yields 10 year event rates that can be translated into Framingham Risk Score equivalents (Table 5.4). CAC >400 is a CAD equivalent,

H.S. Hecht

108 Table 5.5 Reclassification of FRS risk by CAC primary prevention outcome studies Study MESA FRS 0–6 % FRS 6–20 % FRS >20 % NRI Heinz Nixdorf FRS 20 % Rotterdam FRS 20 % NRI

% reclassified

N 5878

Age 62.2

Follow up (years) 5.8

4487

45–75

5.0

2028

69.6

9.2

11.6 % 54.4 % 35.8 % 25 % 15.0 % 65.6 % 34.2 % 12 % 52 % 34 % 19 %

Abbreviations: CAC coronary artery calcium, FRS Framingham risk score, MESA multiethnic study of atherosclerosis

with 10 year event rates exceeding 20 % in asymptomatic patients. The absence of calcified plaque conveys an extraordinarily low 10 year risk (1.1–1.7 %), irrespective of the number of risk factors [56]. Of critical importance is the net reclassification index (NRI) conferred by CAC in the asymptomatic population by three major prospective population based studies [23, 49, 55] (Table 5.5). The percentage of patients with FRS risk estimate correctly reclassified by CAC based on outcomes ranged from 52 to 65.6 % in the intermediate risk population, 34–35.8 % in the high risk group and 11.6–15 % in the low risk cohort, with NRI’s for the entire study population from 19 to 25 %.

Zero Coronary Artery Calcium Scores Individuals with zero CAC scores have not yet developed detectable, calcified coronary plaque but they may have fatty streaking and early stages of plaque. Non-calcified plaques are present in many young adults. Nonetheless, the event rate in patients with CAC score 0 is very low [40, 45, 46]. Raggi et al. [40] demonstrated an annual event rate of 0.11 % in asymptomatic subjects with 0 scores (amounting to a 10-year risk of only 1.1 %), and in the St Francis Heart Study [46], scores of 0 were associated with a 0.12 % annual event rate over the ensuing 4.3 years. Greenland et al. [45], in a higher-risk asymptomatic cohort, noted a higher annual event rate (0.62 %) with 0 CAC scores; a less sensitive CAC detection technique and marked ethnic heterogeneity may have contributed to their findings. In the definitive MESA study [23], 0 CAC was associated with a 0.11 % annual event rate. In a meta-analysis of 64,873 patients followed for

4.2 years [54], the coronary event rate was 0.13 %/year in the 25,903 patients with 0 CAC compared to 1 %/year for the 42,283 with CAC >0. In an analysis of all cause mortality in 44,052 asymptomatic patients followed for 5.6 years [54], the deaths/1000 patient years for the 19,898 with 0 CAC was 0.87, compared to 1.92 for CAC 1–10, and 7.48 for CAC >10. While non-calcified, potentially “vulnerable” plaque is by definition not detected by CAC testing, CAC can identify the pool of higher-risk asymptomatic patients out of which will emerge approximately 95 % of the patients presenting each year with sudden death or an acute myocardial infarction (MI). While the culprit lesion contains calcified plaque in only 80 % of the acute events [57], of greater importance is the observation that exclusively soft, non-calcified plaque has been seen in only 5 % of acute ischemic syndromes in both younger and older populations [12, 58]. In a more recent meta-analysis [38], only 2 of 183 (1.1 %) 0 CAC patients were ultimately diagnosed with an acute coronary syndrome after presenting with acute chest pain, normal troponin, and equivocal EKG findings. CAC >0 had 99 % sensitivity, 57 % specificity, 24 % positive predictive value, and 99 % negative predictive value for ACS. Thus, while it is uncommon that a patient with an imminent acute ischemic syndrome would have had a 0 CAC score, further evaluation, particularly with CCTA, is mandatory.

Adherence to Therapeutic Interventions With the exception of a single study flawed by insufficient power [59], CAC has been shown to have a positive effect on initiation of and adherence to medication and life style changes. In 505 asymptomatic patients, statin adherence 3.6 years after visualizing their CAC scan was 90 % in those with CAC >400 compared to 75 % for 100–399, 63 % for 1–99, and 44 % for 0 CAC (p < 0.0001) [60]. Similarly, in 980 asymptomatic subjects followed for 3 years, ASA initiation, dietary changes, and exercise increased significantly from those with 0 CAC (29 %, 33 %, 44 %, respectively) and was lowest (29 %) in those with CAC >400 (61 %, 67 %, 56 %, respectively [61]. Finally, after a 6 year follow up in 1640 asymptomatic subjects, the odds ratios for those with CAC >0 compared to 0 CAC for usage of statins, ASA, and statin + ASA were 3.53, 3.05 and 6.97, respectively [62]. In the Eisner (Early Identification of Subclinical Atherosclerosis by Noninvasive Imaging Research) trial, 2137 asymptomatic patients were randomized to using CAC to guide treatment or employing usual care [63]. CAC directed care produced significant improvement in systolic blood pressure, LDLC, weight and waist size compared to usual care, without an increase in downstream testing. Patients with CAC >400

5

109

Assessment of Cardiovascular Calcium

Apparently healthy population men >45 year women > 55 year1 Very low risk3

Step 1

Exit

Exit

All >75year receive uncondtional treatment2 . Coronary Artery Calcium Score (CACS) or . Carotid IMT (CIMT) & Carotid Plaque4

Atherosclerosis test

Step 2

Negative test

Positive test

.CACS = 0 . CIMT 65) Off Contrast protocol (370 mgI/cc) Contrast amount (cc) 80–100 Contrast infusion rate (cc/s) 4–5 Saline amount (cc) 50 Saline infusion rate (cc/s) 4–5 Image reconstruction Reconstruction filter Intermediate Slice width (mm) 0.6 Increment (mm) 0.3 mm Matrix 512 × 512 Reconstruction interval Every 5–10 % Image analysis: Axial images, MPR, MIP (cine loops and still frames) Typical scanning protocol for MDCT coronary angiography employed in our institution ECG electrocardiogram, HR heart rate, MPR: multiplanar reformation, MIP maximum intensity projection a If retrospective gating

Fig. 14.7 Axial, non-contrast CT image in a patient with moderate aortic stenosis, demonstrating the quantification of aortic valve calcium (arrow) using the same approach as for coronary calcium scoring. The valvular calcium score (“Agatston”) was 2227

The incremental value of the information derived from the aortic valve calcium score may be particularly useful in patients with low cardiac output and reduced transvalvular gradients. Contrast-enhanced CT can precisely evaluate valve morphology, accurately differentiating trileaflet from bicuspid valves (Fig. 14.8a, b). Planimetric determinations of the aortic valve area (Fig. 14.8c) have shown excellent correlation with echocardiographic and invasive measurements [23–29]. CT has emerged as the integral imaging modality for transcatheter heart valve replacement. As opposed to conventional aortic valve replacement, direct visualization of the valve and annulus is lacking during the TAVR procedure. As a result, imaging is necessary to allow for appropriate valve sizing. CT is used to assess valve morphology, location and degree of calcification, annular sizing, optimum deployment angles, and for presence of peripheral vascular disease. These assessments play a role in predicting success of valve implantation and risk of paravalvular leak in these patients. Expert consensus documents have been released on the use of CT before TAVR stating that CT should be used in the

246

a

N. Shaban et al.

b

c

Fig. 14.8 Double oblique systolic reconstructions of contrastenhanced CT scans showing a tri-leaflet (panel a) and a bicuspid aortic valve (the arrowhead indicates the fusion of the right and left coronary

sinuses; panel b). Planimetry of the valve can be performed subsequently (red contour, panel c): the figure shows a bicuspid aortic valve with moderate stenosis (valve area = 1.2 cm2)

assessment of all patients being considered for TAVR unless contraindicated and that datasets should be interpreted jointly within a multidisciplinary team [30].

commissures, sub-valvular apparatus or even the left atrial wall. MS is often accompanied by marked atrial enlargement involving the appendage. The presence or absence of thrombus in the left atrial appendage can be determined after contrast administration with very high sensitivity although lower specificity since slow flow may impair opacification, which may be increased by adding delayed imaging [34, 35]. Planimetry of mitral valve opening by CT provides accurate assessment of MS severity (Fig. 14.11) [36].

Aortic Regurgitation CT may be useful in evaluating the mechanism leading to aortic regurgitation (AR). AR caused by degenerative valve disease is characterized by thickened and/or calcified leaflets, and the area of lack of coaptation may be visualized in diastolic phase reconstructions centrally or at the commissures. In cases of AR secondary to enlargement of the aortic root, the regurgitant orifice is typically located centrally (Fig. 14.9). Other etiologies that can be depicted include interposition of an intimal flap in cases of dissection, valve distortion or perforation in cases of endocarditis, or leaflet prolapse observed in dissection and in Marfan syndrome. Regurgitant orifice areas measured by planimetry using MDCT correlate well with echocardiographic parameters of AR severity, such as the vena contracta width and the ratio of regurgitant jet to left ventricular outflow tract height, and allow for the detection of moderate and severe AR with high accuracy [31–33].

Mitral Stenosis As in the case of aortic valve calcification, the presence of calcium in the mitral annulus is associated with systemic atherosclerosis and carries negative prognostic implications. The amount of mitral annular calcium can also be quantified with CT (Fig. 14.10), although reproducibility appears to be somewhat lower [18]. In rheumatic mitral stenosis (MS), calcification can extend to the leaflets,

Mitral Regurgitation Both echocardiography and cardiac CT have high sensitivities (92.3 % and 84.6 %, respectively) and specificities (100 % each) for assessing mitral valve abnormalities compared with intraoperative findings, and echocardiography is more sensitive than CT for depicting each prolapsed leaflet of the mitral valve [37]. Echocardiography has been considered the reference imaging modality for mitral valve evaluation given the radiation dose exposure and inferior temporal resolution of CT. In mitral valve prolapse, for example, the use of echocardiography alone to identify the exact site of prolapse is clinician dependent and sometimes difficult, even for those with expertise, because of the limited acoustic window and the complex structure of the mitral apparatus. In patients with mitral valve prolapse, CT can demonstrate the presence of leaflet thickening or the degree and location of prolapse (Fig. 14.12 and Video 14.1). In cases of MR secondary to annular enlargement, often accompanying dilated cardiomyopathy, dimensions of the annulus can be accurately quantified, and a central area of insufficient leaflet coaptation may be observed. Although quantifying MR degree may be difficult, preliminary data suggests that

14 Computed Tomography Evaluation in Valvular Heart Disease

247

Fig. 14.9 Contrast-enhanced MDCT in a patient with an aneurysmal aorta and aortic insufficiency. The valvular plane (yellow line; left lower panel) is oriented perpendicular to two orthogonal planes aligned with

the ascending aorta (red and green lines). A large, central area of insufficient leaflet coaptation during diastole (right lower panel; arrowhead) can be visualized

planimetry of the regurgitant orifice by CT correlates well with echocardiographic grading of severity [38].

to evaluate by echocardiography in the adult patient. Therefore, CT and MRI, due to their good spatiotemporal resolution, large field of view, and multiplanar reconstruction techniques, are playing increasingly important roles in the evaluation of this valve. For visualizing the pulmonary valve, the CT intravenous contrast medium injection protocol should be optimized to ensure that there is adequate contrast opacification in the right cardiac chambers. For morphological evaluation of the valve,

Pulmonic Valve Disease Pathology of the pulmonic valve, whether from idiopathic causes, infective endocarditis, thrombus, regurgitation/ stenosis, or secondary to congenital heart disease is difficult

248

N. Shaban et al.

prospective electrocardiography (ECG) triggered acquisition should be used to minimize radiation dose. However, if functional analysis of the valve or the RV is desired, retrospective ECG-gated multi-phasic acquisition with tube current modulation is the ideal scanning mode [39].

Infective Endocarditis Studies have shown that cardiac-gated CTA has excellent sensitivity, specificity, and positive predictive and negative predictive values in the preoperative evaluation for suspected infective endocarditis, in addition to excellent correlations with preoperative TEE and intraoperative findings [40].

Vegetations are often mobile and tend to be on the atrial aspect of atrioventricular valves, and on the ventricular aspect of semilunar valves (Fig. 14.13). CT can be particularly useful in the demonstration of perivalvular abscesses as fluid-filled collections (Fig. 14.14) that may retain contrast in delayed imaging [41]. In a recent study, MDCT correctly identified 26 out of 27 (96 %) patients with valvular vegetations and 9 out of 9 (100 %) patients with abscesses, which were better characterized by MDCT than with transesophageal echocardiography [42]. Intravascular contrast administration should be optimized, and intravascular attenuation can be further accentuated by the use of 100-kV scan protocols whenever possible. Although the maximal temporal resolution of a scanner cannot be altered, the reconstruction frame of the dataset can and should be optimized when assessing valvular function. Reconstruction of 20- or 25-phase datasets (at 5 % or 4 % increments of the R-R interval) provides improved temporal depiction of valve motion that facilitates cine evaluation of valvular pathology, such as hypermobile vegetations. In addition, advanced image processing techniques, such as blood pool inversion (BPI) volume-rendering, can be used to allow 3-Dimensional/4-Dimensional (3-D/4-D) assessment of valvular structure and function [43]. In patients with aortic valve endocarditis with highly mobile vegetations, CT may be especially attractive as an alternative to invasive coronary angiography for preoperative evaluation.

Prosthetic Valves

Fig. 14.10 Short-axis view at the level of the mitral valve, showing extensive annular calcification (arrows)

a

b

Fig. 14.11 Contrast-enhanced CT scan in the four-chamber and shortaxis views (panels a and b, respectively) from a patient with rheumatic mitral stenosis. The typical thickening and restricted dome-shaped

Many of the aforementioned features of native VHD apply also to the evaluation of cardiac bioprostheses. Transthoracic echocardiography is useful for prosthetic valve evaluation, but can be limited by acoustic shadowing and poor acoustic windowing. Recently, cardiac CT has been recognized as a viable alternative to evaluation of prosthetic valve complications

c

opening of the leaflets can be observed (arrows and asterisk). Planimetry of the valve (panel c) demonstrated moderate stenosis (red contour; area = 1.3 cm2)

14 Computed Tomography Evaluation in Valvular Heart Disease

including valve thrombosis, dehiscence, pannus development, endocarditis, and paravalvular leak. However, careful attention to CT technique, achieving prescan target heart rates, extensive windowing adjustments, and awareness of normal postoperative paravalvular structures is imperative. Some valves, such as ball in cage valves, are not readily evaluable by CT because of extreme beam hardening artifact from the thicker metal struts found in these models. Whereas evaluation of most other valves using a very soft window with consider-

249

able windowing adjustments is to minimize beam hardening is certainly possible [44]. Recent work suggests iterative reconstructions may reduce beam-hardening artifact from prosthetic valves compared with filtered back projection reconstruction techniques [45]. Motion artifact can be adequately reduced by administration of beta-blockade to achieve heart rates between 50 and 60 bpm. Motion artifact is worst for aortic valve prosthesis during ventricular systole and for mitral valve prosthesis during end-diastole. Thus, it has been found that imaging in mid-diastole is the most ideal for prosthetic valve evaluation [46]. CT is particularly useful for the evaluation of some types of mechanical valves. In Prostheses with two discs should open symmetrically (Fig. 14.15 and Video 14.2). In those with a single disc, the angle of opening can also be measured [47]. Finally, heterografts and homografts can be evaluated completely, including the distal anastomosis and the patency of the coronary arteries if these were reimplanted.

Imaging Pearls

Fig. 14.12 End-systolic three-chamber view of the left ventricle demonstrating prolapse of the posterior mitral leaflet (arrow)

a

Fig. 14.13 Diastolic (panel a) and systolic (panel b) reconstructions of a contrast-enhanced MDCT study in a patient with a bioprosthesis in the aortic position. A large, mobile vegetation that prolapses into the

• Plan ahead; as this will allow for imaging protocol optimization if valvular evaluation will be attempted. • If simultaneous assessment of the right heart structures is intended, the contrast protocol should be optimized. An initial bolus of 80–100 cc followed by a mixture of

b

ascending aorta in systole can be noted (black arrows). In addition, perivalvular thickening and fluid-filled collections can be noted (white arrows) indicating the presence of a perivalular abscess

250

N. Shaban et al.

a

b

c

d

Fig. 14.14 Evaluation of mechanical prostheses by CT. The top row shows contrast-enhanced images (systole, panel a; diastole, panel b) of a normal-functioning mechanical prosthesis in the mitral position. The two discs close and open completely and symmetrically (white arrows) during the cardiac cycle. Comparable systolic (panel c) and diastolic

(panel d) reconstructions of a non-contrast CT evaluation of a dysfunctional mitral prosthesis. One of the discs does not open in diastole (white arrowhead). Subsequent surgical intervention demonstrated prosthetic thrombosis

contrast and saline (1:1) at 4–5 cc/s will result in adequate coronary evaluation and sufficient right-heart opacification without excessive enhancement. Alternatively, a second infusion of contrast administered at a slower rate (2–3 cc/s) can be employed [15, 16]. • Quantification of ventricular end-systolic volumes and the degree of MR and AS requires adequate image quality during systole. It may be necessary to avoid tube current modulation in these cases. Alternatively, the maximal

tube output can be timed to end-systole, which will provide adequate depiction of mitral closure and aortic opening, as well as potentially motionless coronary images, particularly at higher heart rates. • If the entire thoracic aorta needs to be imaged (i.e. in cases of aneurysm with associated AR) and coronary evaluation is not required, using thicker detector collimation will enable reductions in radiation dose and breath-hold duration.

14 Computed Tomography Evaluation in Valvular Heart Disease

251

• •

•

Fig. 14.15 Contrast-enhanced CT in a patient with pulmonary infundibular stenosis (arrowheads). The contrast protocol was optimized to provide adequate opacification of right heart chambers

•

• • Most patients with VHD can tolerate beta-blockers for optimal coronary evaluation. However, caution and smaller doses are recommended in cases with severe degrees of left ventricular dysfunction/dilatation, AS, AR or pulmonary hypertension. • Atrial fibrillation is common in patients with VHD. It may lead to a decrease in image quality and accuracy of valvular and ventricular assessment, although this is typically more significant for evaluation of the coronary arteries. • For the evaluation of ventricular or valvular function with MDCT, reconstructions at every 10 % of the RR interval are usually sufficient. In specific cases, a more detailed evaluation of the valve can be obtained by reconstructing images at smaller intervals (i.e. every 5 %) in the cardiac phase of interest (for example, during systole for AS) [48]. • The combination of cine loops and still frames facilitates the detection of valvular abnormalities. • CT imaging in the evaluation for TAVR should include imaging of the aortic root, aorta, and iliac, as well as common femoral arteries. To achieve the desired accuracy and to allow for adequate motion-free images, imaging of the aortic root must be synchronized to the electrocardiogram (ECG) either by retrospective ECG gating or by prospective ECG triggering, depending on patient characteristics. It is not necessary to image the

•

•

•

entire aorta and iliofemoral arteries with ECG synchronization. For these sections, non-gated acquisitions will allow lower radiation exposure and faster volume coverage requiring lower contrast volumes. Since detailed dimensions of the aortic root and of the iliofemoral arteries must be obtained, spatial resolution must be high enough to provide adequate imaging. The optimal acquisition protocol is that which obtains a reconstructed slice width of 3700 has a positive predictive value of near 100 % [25]. The optimal plane to perform planimetry of the valvular area is parallel to the annulus as determined from two orthogonal double-oblique views perpendicular to the valve plane. The optimal level of that plane is the one showing the smallest area during the phase of maximum valve opening (Fig. 14.9). Quantification of the regurgitant volume/fraction from the difference in right and left stroke volumes is only accurate for isolated regurgitant lesions. A score evaluating leaflet mobility and thickening, subvalvular thickening and calcification, as well as the presence of left atrial thrombus, may determine whether MS can be treated percutaneously or surgically. CT can provide useful information for all of these features. The mitral valve is divided into the anterolateral commissure, posteromedial commissure, anterior leaflet and posterior leaflet. The leaflets are subdivided into three segments each (A1, A2 and A3; and P1, P2, and P3, from lateral to medial). Determination of which segments are affected and to what degree determines in part the likelihood of successful surgical repair in mitral valve prolapse. Sharper reconstruction filters and increasing window level of the image display facilitates evaluation of mechanical prosthetic valves. Optimum valve evaluation for both aortic and mitral prosthetic valves is best achieved during mid-diastole.

References 1. Nkomo VT, Gardin JM, Skelton TN, Gottdiener JS, Scott CG, Enriquez-Sarano M. Burden of valvular heart diseases: a populationbased study. Lancet. 2006;368(9540):1005–11. 2. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd, Guyton RA, O’Gara PT, Ruiz CE, Skubas NJ, Sorajja P, Sundt TM 3rd, Thomas JD. 2014 AHA/ACC guideline for the management of

252

3.

4.

5.

6.

7.

8.

9. 10.

11.

12.

13.

14.

15.

N. Shaban et al. patients with valvular heart disease: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63(22): 2438–88. Bonow RO, Carabello BA, Chatterjee K, de Leon Jr AC, Faxon DP, Freed MD, et al. 2008 Focused update incorporated into the ACC/ AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients With Valvular Heart Disease): endorsed by the Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation. 2008;118(15):e523–661. Cawley PJ, Maki JH, Otto CM. Cardiovascular magnetic resonance imaging for valvular heart disease: technique and validation. Circulation. 2009;119(3):468–78. Bennett CJ, Maleszewski JJ, Araoz PA. CT and MR imaging of the aortic valve: radiologic-pathologic correlation. Radiographics Rev Publ Radiol Soc North Am Inc. 2012;32(5):1399–420. Taylor AJ, Cerqueira M, Hodgson JM, Mark D, Min J, O’Gara P, et al. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 appropriate use criteria for cardiac computed tomography. A report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol. 2010;56(22):1864–94. Orakzai SH, Orakzai RH, Nasir K, Budoff MJ. Assessment of cardiac function using multidetector row computed tomography. J Comput Assist Tomogr. 2006;30(4):555–63. Reiter SJ, Rumberger JA, Stanford W, Marcus ML. Quantitative determination of aortic regurgitant volumes in dogs by ultrafast computed tomography. Circulation. 1987;76(3):728–35. Boxt LM. CT of valvular heart disease. Int J Cardiovasc Imaging. 2005;21(1):105–13. Gilard M, Cornily JC, Pennec PY, Joret C, Le Gal G, Mansourati J, et al. Accuracy of multislice computed tomography in the preoperative assessment of coronary disease in patients with aortic valve stenosis. J Am Coll Cardiol. 2006;47(10):2020–4. Meijboom WB, Mollet NR, Van Mieghem CA, Kluin J, Weustink AC, Pugliese F, et al. Pre-operative computed tomography coronary angiography to detect significant coronary artery disease in patients referred for cardiac valve surgery. J Am Coll Cardiol. 2006;48(8):1658–65. Reant P, Brunot S, Lafitte S, Serri K, Leroux L, Corneloup O, et al. Predictive value of noninvasive coronary angiography with multidetector computed tomography to detect significant coronary stenosis before valve surgery. Am J Cardiol. 2006;97(10):1506–10. Scheffel H, Leschka S, Plass A, Vachenauer R, Gaemperli O, Garzoli E, et al. Accuracy of 64-slice computed tomography for the preoperative detection of coronary artery disease in patients with chronic aortic regurgitation. Am J Cardiol. 2007;100(4):701–6. Russo V, Gostoli V, Lovato L, Montalti M, Marzocchi A, Gavelli G, et al. Clinical value of multidetector CT coronary angiography as a preoperative screening test before non-coronary cardiac surgery. Heart (Br Cardiac Soc). 2007;93(12):1591–8. Litmanovich D, Zamboni GA, Hauser TH, Lin PJ, Clouse ME, Raptopoulos V. ECG-gated chest CT angiography with 64-MDCT and tri-phasic IV contrast administration regimen in patients with acute non-specific chest pain. Eur Radiol. 2008;18(2):308–17.

16. Takakuwa KM, Halpern EJ. Evaluation of a “triple rule-out” coronary CT angiography protocol: use of 64-Section CT in low-tomoderate risk emergency department patients suspected of having acute coronary syndrome. Radiology. 2008;248(2):438–46. 17. Rosenhek R, Binder T, Porenta G, Lang I, Christ G, Schemper M, et al. Predictors of outcome in severe, asymptomatic aortic stenosis. N Engl J Med. 2000;343(9):611–7. 18. Budoff MJ, Takasu J, Katz R, Mao S, Shavelle DM, O’Brien KD, et al. Reproducibility of CT measurements of aortic valve calcification, mitral annulus calcification, and aortic wall calcification in the multi-ethnic study of atherosclerosis. Acad Radiol. 2006;13(2):166–72. 19. Koos R, Kuhl HP, Muhlenbruch G, Wildberger JE, Gunther RW, Mahnken AH. Prevalence and clinical importance of aortic valve calcification detected incidentally on CT scans: comparison with echocardiography. Radiology. 2006;241(1):76–82. 20. Messika-Zeitoun D, Aubry MC, Detaint D, Bielak LF, Peyser PA, Sheedy PF, et al. Evaluation and clinical implications of aortic valve calcification measured by electron-beam computed tomography. Circulation. 2004;110(3):356–62. 21. Koos R, Mahnken AH, Sinha AM, Wildberger JE, Hoffmann R, Kuhl HP. Aortic valve calcification as a marker for aortic stenosis severity: assessment on 16-MDCT. AJR Am J Roentgenol. 2004;183(6):1813–8. 22. Shavelle DM, Budoff MJ, Buljubasic N, Wu AH, Takasu J, Rosales J, et al. Usefulness of aortic valve calcium scores by electron beam computed tomography as a marker for aortic stenosis. Am J Cardiol. 2003;92(3):349–53. 23. Alkadhi H, Wildermuth S, Plass A, Bettex D, Baumert B, Leschka S, et al. Aortic stenosis: comparative evaluation of 16-detector row CT and echocardiography. Radiology. 2006;240(1):47–55. 24. Bouvier E, Logeart D, Sablayrolles JL, Feignoux J, Scheuble C, Touche T, et al. Diagnosis of aortic valvular stenosis by multislice cardiac computed tomography. Eur Heart J. 2006;27(24):3033–8. 25. Cowell SJ, Newby DE, Burton J, White A, Northridge DB, Boon NA, et al. Aortic valve calcification on computed tomography predicts the severity of aortic stenosis. Clin Radiol. 2003;58(9):712–6. 26. Feuchtner GM, Dichtl W, Friedrich GJ, Frick M, Alber H, Schachner T, et al. Multislice computed tomography for detection of patients with aortic valve stenosis and quantification of severity. J Am Coll Cardiol. 2006;47(7):1410–7. 27. Feuchtner GM, Muller S, Bonatti J, Schachner T, Velik-Salchner C, Pachinger O, et al. Sixty-four slice CT evaluation of aortic stenosis using planimetry of the aortic valve area. AJR Am J Roentgenol. 2007;189(1):197–203. 28. LaBounty TM, Sundaram B, Agarwal P, Armstrong WA, Kazerooni EA, Yamada E. Aortic valve area on 64-MDCT correlates with transesophageal echocardiography in aortic stenosis. AJR Am J Roentgenol. 2008;191(6):1652–8. 29. Lembcke A, Thiele H, Lachnitt A, Enzweiler CN, Wagner M, Hein PA, et al. Precision of forty slice spiral computed tomography for quantifying aortic valve stenosis: comparison with echocardiography and validation against cardiac catheterization. Invest Radiol. 2008;43(10):719–28. 30. Achenbach S, Delgado V, Hausleiter J, Schoenhagen P, Min JK, Leipsic JA. SCCT expert consensus document on computed tomography imaging before transcatheter aortic valve implantation (TAVI)/transcatheter aortic valve replacement (TAVR). J Cardiovasc Comput Tomogr. 2012;6(6):366–80. 31. Alkadhi H, Desbiolles L, Husmann L, Plass A, Leschka S, Scheffel H, et al. Aortic regurgitation: assessment with 64-section CT. Radiology. 2007;245(1):111–21. 32. Feuchtner GM, Dichtl W, Muller S, Jodocy D, Schachner T, Klauser A, et al. 64-MDCT for diagnosis of aortic regurgitation in patients

14 Computed Tomography Evaluation in Valvular Heart Disease

33.

34.

35.

36.

37.

38.

39. 40.

41.

referred to CT coronary angiography. AJR Am J Roentgenol. 2008;191(1):W1–7. Jassal DS, Shapiro MD, Neilan TG, Chaithiraphan V, Ferencik M, Teague SD, et al. 64-slice multidetector computed tomography (MDCT) for detection of aortic regurgitation and quantification of severity. Invest Radiol. 2007;42(7):507–12. Kim YY, Klein AL, Halliburton SS, Popovic ZB, Kuzmiak SA, Sola S, et al. Left atrial appendage filling defects identified by multidetector computed tomography in patients undergoing radiofrequency pulmonary vein antral isolation: a comparison with transesophageal echocardiography. Am Heart J. 2007;154(6):1199–205. Hur J, Kim YJ, Lee HJ, Ha JW, Heo JH, Choi EY, et al. Left atrial appendage thrombi in stroke patients: detection with two-phase cardiac CT angiography versus transesophageal echocardiography. Radiology. 2009;251(3):683–90. Messika-Zeitoun D, Serfaty JM, Laissy JP, Berhili M, Brochet E, Iung B, et al. Assessment of the mitral valve area in patients with mitral stenosis by multislice computed tomography. J Am Coll Cardiol. 2006;48(2):411–3. Ghosh N, Al-Shehri H, Chan K, Mesana T, Chan V, Chen L, et al. Characterization of mitral valve prolapse with cardiac computed tomography: comparison to echocardiographic and intraoperative findings. Int J Cardiovasc Imaging. 2012;28(4):855–63. Alkadhi H, Wildermuth S, Bettex DA, Plass A, Baumert B, Leschka S, et al. Mitral regurgitation: quantification with 16-detector row CT--initial experience. Radiology. 2006;238(2):454–63. Rajiah P, Nazarian J, Vogelius E, Gilkeson RC. CT and MRI of pulmonary valvular abnormalities. Clin Radiol. 2014;69(6):630–8. Gahide G, Bommart S, Demaria R, Sportouch C, Dambia H, Albat B, et al. Preoperative evaluation in aortic endocarditis: findings on cardiac CT. AJR Am J Roentgenol. 2010;194(3):574–8. Gilkeson RC, Markowitz AH, Balgude A, Sachs PB. MDCT evaluation of aortic valvular disease. AJR Am J Roentgenol. 2006; 186(2):350–60.

253 42. Feuchtner GM, Stolzmann P, Dichtl W, Schertler T, Bonatti J, Scheffel H, et al. Multislice computed tomography in infective endocarditis: comparison with transesophageal echocardiography and intraoperative findings. J Am Coll Cardiol. 2009;53(5): 436–44. 43. Entrikin DW, Gupta P, Kon ND, Carr JJ. Imaging of infective endocarditis with cardiac CT angiography. J Cardiovasc Comput Tomogr. 2012;6(6):399–405. 44. O’Neill AC, Martos R, Murtagh G, Ryan ER, McCreery C, Keane D, et al. Practical tips and tricks for assessing prosthetic valves and detecting paravalvular regurgitation using cardiac CT. J Cardiovasc Comput Tomogr. 2014;8(4):323–7. 45. Sucha D, Willemink MJ, de Jong PA, Schilham AM, Leiner T, Symersky P, et al. The impact of a new model-based iterative reconstruction algorithm on prosthetic heart valve related artifacts at reduced radiation dose MDCT. Int J Cardiovasc Imaging. 2014;30(4):785–93. 46. Symersky P, Budde RP, Westers P, de Mol BA, Prokop M. Multidetector CT imaging of mechanical prosthetic heart valves: quantification of artifacts with a pulsatile in-vitro model. Eur Radiol. 2011;21(10):2103–10. 47. Konen E, Goitein O, Feinberg MS, Eshet Y, Raanani E, Rimon U, et al. The role of ECG-gated MDCT in the evaluation of aortic and mitral mechanical valves: initial experience. AJR Am J Roentgenol. 2008;191(1):26–31. 48. Abbara S, Pena AJ, Maurovich-Horvat P, Butler J, Sosnovik DE, Lembcke A, et al. Feasibility and optimization of aortic valve planimetry with MDCT. AJR Am J Roentgenol. 2007;188(2):356–60. 49. Ruhl KM, Das M, Koos R, Muhlenbruch G, Flohr TG, Wildberger JE, et al. Variability of aortic valve calcification measurement with multislice spiral computed tomography. Invest Radiol. 2006;41(4):370–3.

Transcatheter Aortic Valve Implantation (TAVI)

15

Chesnal Dey Arepalli, Christopher Naoum, Philipp Blanke, and Jonathon A. Leipsic

Abstract

Computerized tomography (CT) plays a pivotal role in selection of patients’, appropriate device size and pre-procedural guidance in successful outcome of transcatheter aortic valve implantation (TAVI). CT based vascular and non-vascular evaluation and integration of relevant measurements into TAVI work up has been shown to reduce morbidity and mortality. Post-procedural device assessment and complications could also be reliably evaluated with CT. Utilization of CT is not just confined to TAVI but it is also increasingly being used for any transcatheter valvular assessment. Keywords

AS – Aortic Stenosis • AVC – Aortic Valvular Calcifications • AA – Aortic Annulus • SOV – Sinus of Valsalva • Coronary Artery Ostium • PAR – Paravalvular Regurgitation • CT – Computerized tomography • TTE – Transthoracic Echocardiography • TEE – Transesophageal Echocardiography • TAVI – Transcatheter Aortic Valve Implantation

Introduction Transcatheter aortic valve implantation (TAVI) is an alternative treatment option for symptomatic severe aortic stenosis (AS) patients who are deemed inoperable or at high risk for surgical aortic valve replacement (AVR). TAVI as the name suggests is a procedure where in a bio-prosthetic device is implanted without removing the diseased valve via a transcatheter approach. C.D. Arepalli, MBBS, DNB • C. Naoum, MBBS, FRACP Department of Radiology, St Paul’s Hospital UBC, Vancouver, BC, Canada P. Blanke, MD Department of Medicine, St Paul’s Hospital UBC, Vancouver, BC, Canada J.A. Leipsic, MD, FRCPC, FSCCT (*) Department of Radiology, St. Paul’s Hospital, Room P2214 – 2nd Floor Providence Building 1081 Burrard Street, Vancouver, BC V6Z 1Y6, Canada e-mail: [email protected]

The most common cause for AS is age related progressive calcification of a normal aortic valve (Fig. 15.1a–c) [1, 2]. Additionally, congenital causes like bicuspid aortic valve or inflammatory conditions like rheumatic heart disease are also common causes of AS [1]. The prevalence of senile AS is increasing as the mean age of Western societies is increasing [3, 4]. Current data has documented the prevalence of AS in the elderly population (age ≥ 75 years) as 12.4 % and severe AS is 3.4 % [3]. Although senile or degenerative aortic stenosis occurs primarily in the elderly population, with mean age of 65–70 years, congenital bicuspid aortic valve (BAV) with significant aortic stenosis tends to occur slightly earlier between 15 and 65 years of age (Fig. 15.1d) [5, 6]. In one study it was observed that BAV was the commonest cause of AS between the ages of 60 and 75 years (59 % of cases) and those aged less than 60 years, BAV was a causative factor in 40 % [7]. BAV AS were found to require AVR 5 years before than those with a tricuspid valve [8]. Importantly, the natural clinical progression of symptomatic AS is rapid and associated with a poor clinical

© Springer International Publishing 2016 M.J. Budoff, J.S. Shinbane (eds.), Cardiac CT Imaging: Diagnosis of Cardiovascular Disease, DOI 10.1007/978-3-319-28219-0_15

255

256

C.D. Arepalli et al.

a

b

c

d

Fig. 15.1 Aortic valvular calcification (AVC). AVC of varying degrees – (a) mild, (b) moderate, (c) severe, (d) Bicuspid aortic valve (BAV) with calcification. Note the asymmetry of the cusps

with a larger non-coronary cusp and raphae between the right and the left coronary cusps in keeping with a Type 1 A Bicuspid Valve

outcome with an average time of death shown to be within 5 years after onset of angina, 3 years after onset of syncope and within 1–2 years after onset of heart failure symptoms [9]. Until the advent of TAVI, the sole effective treatment option for symptomatic AS was surgical AVR. Medical treatment alone had very poor prognosis with a mortality rate of 50 % at 2 years. Unfortunately, many patients with severe symptomatic AS are denied this surgery owing to associated

co-morbidities in the elderly population that make surgical risk prohibitive [3, 10]. TAVI is a new alternate therapy that has been shown through many registries, meta-analyses, and a number of randomized trials to be an effective therapy for severe symptomatic AS who were deemed inoperable and high risk surgical patients [11, 12]. TAVI was first pioneered by Cribier et al. in 2002 through transvenous transeptal approach [13]. However, over the past decade, alternative procedures have

15 Transcatheter Aortic Valve Implantation (TAVI)

been developed and at present retrograde transarterial transfemoral path, first described by Webb and colleagues in 2005, is the most favored approach. As opposed to surgical AVR where it is possible to directly visualize and measure the size of the aortic root; TAVI relies indirectly on imaging measurements before implantation. Based on these measurements; device selection and the procedural approach are planned. Pre-procedural imaging plays a critical role in ensuring the success of the procedure and also to minimize the peri and post-procedural complications.

Imaging Modalities Historically, TAVI imaging was reliant on measurements derived form 2-dimensional echocardiography and invasive angiography. However, multidetector computed tomography (MDCT) has grown to become an essential tool for the assessment of aortic root/annulus, evaluation of thoracic and abdominal aorta and ilio-femoral vessels. Moreover, MDCT is being used to predict optimal co-planar projection angles for device deployment. Current state of the art CT scanners acquire data volumetrically with isotropic spatial and high temporal resolution. Post processing of data provide multiplanar and curved planar reformations (MPR and CPR), volume rendered technique images (VRT), minimum and maximum intensity projections (MIPmin and MIPmax respectively) which are essential to accurately size the aortic annulus, aortic root and other vascular access sites. Integration of CT data into sizing algorithms and work flow has consistently demonstrated significant reduction in the incidence of paravalvular and vascular complications [14–16]. Thus CT has become an accepted integral part in the overall work flow of TAVI patients. Of late, CT has also been shown to play an important additional role in valve in valve procedures. The role of MDCT for valvular intervention can be broadly discussed under three headings: (a) Pre-procedural, (b) Peri-procedural and (c) Post procedural. Although the focus of this chapter is on pre-procedural role of CT; it is equally pertinent to understand the peri and post procedural complications as they play an important role in understanding the dynamics of TAVI and thus indirectly influence the pre-procedural work up. Before one can proceed with CT based ‘Sizing of the aortic annulus’ and procedural planning in the work up of TAVI patients, it is important to understand the complex aortic valve and root anatomy. Knowledge of the 3 dimensional (3 D) complex aortic root geometry and consideration of

257

various anatomical measurements are critical in the determination of patient eligibility, bioprosthesis size, procedural planning and to anticipate procedural complications.

Anatomy Aortic Root The Aortic root is a further extension of the left ventricular outflow tract (LVOT). As a result, the LVOT should be considered an anatomical segment between the left ventricle and ascending aorta. As the aortic root is a complex dynamic structure; understanding the morphology and kinesis of the aortic root subcomponents helps in better appreciation of interplay of different sub-components with each other and also with adjacent structures. This not only helps in procedural success but also has the potential in refinement of the bio-prosthesis devices. Aortic root sub-components include aortic annulus (AA), aortic sinus and the valvular leaflets, the coronary ostia and the sinotubular junction.

Aortic Annulus Broadly, two terms define the aortic annulus – (a) anatomic annulus and (b) virtual basal ring. The Anatomical Aortic annulus is more of a histological demarcation that is defined as a fibrous structure that attaches the aortic root to the left ventricle. It consists of three coronets like structures that support the valvular leaflets. The Virtual basal annulus is defined by nadir (lowest depth) plane of each of the coronets and it corresponds to the aortic-ventricular junction. Although the term annulus suggests a circular shape, with the use of 3-D MDCT imaging , it has been found to be fairly consistently to be of a non-circular geometry with often elliptical and sometimes oval shaped configuration [17, 18]. It has been also observed that there are considerable variations in shape, size and direction of the annulus during the cardiac cycle (Fig. 15.2). These are influenced by deformation, stretch, compliance of the aortic root as well as left ventricular geometry, mitral valve anatomy and pathology [19, 20]. Further, Nakai et al. [21] showed that there is cranial displacement of aortic annulus during early systole and in diastole whereas in rest of the systole and in isovolumetric relaxation there is caudal displacement. It has been observed in multiple studies, that the aortic annulus measurements in cohorts without and with aortic stenosis show largest size (area) and diameter in systole and smallest measurements during diastole due to cyclical changes [22–26].

258

C.D. Arepalli et al.

Imaging Modalities and Aortic Annulus Assessment Strengths and Limitations of Echocardiography and MDCT Accurate and reproducible pre-operative measurements of annulus are of paramount importance that helps in patient selection, device selection and size for TAVI procedure and also forecasting the outcome of the geometrical configuration of aortic annulus-prosthesis after implantation. In addition, better understanding of the aortic root complex and precise measurements would help in predicting and mitigating peri and post procedural complications. Owing to the complex 3D aortic root geometry, measurements in a single plane have been shown to lead to both under or overestimation (depending upon the plane used) and have therefore been shown to be limited value for TAVI sizing. Two dimensional (2D) measurements as provided by transthoracic echocardiography (TTE) and transesophageal

echocardiography (TEE) are routinely used. In fact, the current TAVI device sizes provided by vendors are based on 2D echocardiography measurements. However, it was noted in several studies that echocardiography measurements underestimated the true aortic annular size and were smaller in size when compared to MDCT. In a meta-analysis, MDCT aortic annulus diameter measurements on coronal view were 25.3 ± 0.52 mm which were larger than sagittal view measurements by MDCT (22.7 ± 0.37 mm), TTE (22.6 ± 0.28 mm), and TEE (23.1 ± 0.32 mm) [27]. In a study by Mizia-Stec at el, the mean aortic annulus diameter on TTE was 24 ± 3.6 mm, 26 ± 4.2 mm using TEE, and 26.9 ± 3.2 mm on MDCT (P = 0.04 vs. TTE) [28]. Messika-Zeitoun and colleagues [29] observed larger differences between CT and TTE (1.22 ± 1.3 mm) or TEE (1.52 ± 1.1 mm) than the difference between TTE and TEE (0.6 ± 0.8 mm; and p < 0.0001, respectively p = 0.03). Early experiences with the integration of CT not only documented differences in annular measurements between

a

b

c

d

e

f

Fig. 15.2 Dynamic nature of Aortic Annulus over the course of cardiac cycle (75–25 %) (a–f) and corresponding measurements (g–l). Note the variation in the shape and size of the aortic annulus from diastole (65–75 %) through systole (25–35 %). Measurement of aortic annulus is usually performed in systole. The Geometrical

shape of the aortic annulus changes from elliptical in diastole (75 %) to near circular in systole (25 %). The perimeter size increased from 74 to 79 mm (~5 mm difference – ~6 %) variation, whereas area size increased from 4.24 to 4.84 cm2 (~0.6 cm2 difference – ~variation of 13 %)

15 Transcatheter Aortic Valve Implantation (TAVI)

259

g

h

i

j

k

l

Fig. 15.2 (continued)

MDCT and echocardiography but also differences in potential sizing recommendations. Koos et al. evaluated the impact of 3D imaging methods (Cardiac magnetic resonance – CMR and dual source CT – DSCT) versus 2D TEE methods on TAVI selection [30]. In their cohort of 58 patients, different measurements would have changed TAVI strategy in 22–24 % patients. DSCT coronal measurements when compared to 2D TEE would have changed strategy in 16 patients. Further in 14/16 of those patients, the prosthesis size would have been larger and in the rest of the two, it was too large to implant. In a related similar study of 45 patients comparing 2D TTE and TEE and MSCT measurements, Messika-Zeitoun et al. observed that a decision based on MSCT would have modified TAVI strategy in 40–42 % [29]. The limitations of 2-D imaging largely reflect the complex almost uniformly non-circular configuration of the annulus. It is not that 2-D diameters derived from echocardiography or MSCT are inaccurate but it is simply a geometrical reality that the complex morphology of a non-circular structure cannot be accurately characterized with a single 2-D measurement. The growing appreciation of the limitations of 2-D imaging has driven and increased focus on the utilization of

3D modalities in measurement of aortic root including aortic annulus and left ventricular outflow tract. Ng et al. evaluated 2D circular, 3D circular and 3D planimetered annular and LVOT areas by TEE and compared with MSCT planimetered areas [23]. In their cohort, the mean annular area by MSCT planimetry was 4.65 ± 0.82 cm2 whereas by 2D TEE circular (3.89 ± 0.74 cm2, P < 0.001), 3D TEE circular (4.06 ± 0.79 cm2, P < 0.001), and 3D TEE planimetered annular areas (4.22 ± 0.77 cm2, P < 0.001). They observed 3D TEE derived planimetered annular areas had the narrowest limits of agreement and least bias when compared to MDCT. Although the circular geometric assumptions made by 2D and 3D circular measurements were overcome by 3D TEE; Ng et al. observed 3D TEE when compared to MDCT still underestimated annular areas by up to 10 %. MSCT has also been shown to be highly discriminatory of those patients that historically experienced paravalvular regurgitations owing to undersizing secondary to device sizing on the basis of 2-dimensional echocardiography. Willson and colleagues noted that the annular area in particular on MDCT was highly discriminatory of PAR [31]. This was important knowledge as it has been well established the

260

paravalvular regurgitation is associated with increased morbidity and mortality [28–30, 32, 33]. Building upon this retrospective knowledge multiple groups have shown that the integration of MDCT in device selection allows for the reduction in the burden of PAR. Jilaihawi et al, prospective integration of CT guided algorithm reduced worse than mild paravalvular regurgitation (PAR) from 21.9 to 7.5 % when compared to 2D TEE [24]. Building upon this single center data Binder et al. in a prospective, multicenter controlled trial, evaluated the effects of the integration of an area based sizing algorithm on the clinical outcomes post TAVI [34]. With a non-randomized trial design half of the subjects (n = 133) underwent TAVI with the integration of MDCT measurements and sizing recommendations and the other half underwent TAVI without the integration of MDCT. They observed more than mild PAR (primary endpoint) was present in 5.3 % (7 of 133) of the MDCT group and in 12.8 % (17 of 133) in the control group (p = 0.032). The combined secondary endpoint (composite of in-hospital death, aortic annulus rupture, and severe PAR) occurred in 3.8 % (5 of 133) of the MDCT group and in 11.3 % (15 of 133) of the control group (p = 0.02). Finally the recently published large multicenter high risk CoreValve (Medtronic, Minneapolis, MN) trial showed excellent clinical outcomes as compared to surgical aortic valve replacement with the THV selection supported almost exclusively by MDCT perimeter measurements [35].

Dynamism of the Aortic Annulus It is to be noted that aortic annulus is a dynamic structure with variation during systole and diastole phases of the cardiac cycle. The aortic annulus is often elliptical and assumes more circular shape during systole (Fig. 15.2). Further, it has been observed that there is increased asymmetrical deformation during diastole which especially affects the right coronary cusp portion of the annulus [36]. The ellipticity index (EI) defined as ratio of maximum to minimum diameters substantially decreased from diastole to systole due to significant increase in antero-posterior minor diameter [25]. Significant changes in area and radius were observed in both non-diseased annulus and stenotic annulus [37]. This has been attributed to annular reshaping than stretch with corresponding increase in systolic area [25]. Although cross-sectional area (CSA) derived measurements varied during cardiac cycle, it was found perimeter based diameter measurements show negligible increase in patients with calcified valves (0.56 ± 0.85 %; p < 0.001) and very small changes in normal subjects (2.2 ± 2.2 %, p = 0.01) [25]. In another study by Aspern et al., no significant difference was observed with perimeter-derived effective diameters (ED) (mean difference: 0.2 ± 0.4; p = 0.07), however,

C.D. Arepalli et al.

area derived ED showed a significant mean difference (0.4 ± 0.6 mm; p = 0.009), thus reconfirming influence of the cardiac cycle on aortic annulus area measurements [38]. It is likely that the diseased valves/tissues leave little scope for stretch and therefore perimeter derived aortic annulus measurements are least subjected to variation in cardiac cycle (Fig. 15.2). However, it is being noted smoothing algorithms used for perimeter derived measurements are inconsistent across work station platforms and might cause measurement errors. Although area based ED measurements are smaller than perimeter derived ED, both approaches showed good agreement [38]. Whether perimeter or area is chosen for transcatheter heart valve (THV) selection it is essential to understand the impact that the geometrical variable (area or perimeter derived measurements) may have on the prosthesis size that would be selected. Blanke et al. showed assuming a perfect circle with ellipticity index of 1, with increase in nominal diameter, area increases exponentially whereas perimeter increases proportionally [39]. Thus, for e.g. 10 % diameter oversizing would translate to 10 % increase in perimeter whereas it would increase area by 21 %. These geometrical truths are essential to understand and of clinical importance when certain degree of annular oversizing is contemplated during TAVI. In summation, 3D MDCT measurements play a critical role in annular sizing and THV selection. MDCT measurements are accurate, reproducible and choice of area or perimeter derived measurements need to be tailored depending upon the choice of prosthesis devices, allowing for dynamism of the aortic annulus and based on the available tools (software) available at the sites.

MDCT Measurement of the Aortic Annulus Approach to Aortic Annulus Measurement We recommend using the phase of the cardiac cycle with best image quality and in systolic phases (25–45 %) to allow for consistent assessment of the annulus when it is the largest. Aortic annulus measurements are performed orthogonally in relation to the plane of 3 hinge points on a dedicated work station. Manual multiplanar reformats of the annulus is preferred as it helps in better understanding of the annulus and the annular plane rather than relying on automated tools which should ideally only be used by highly experienced operators. Briefly, the steps to achieve ‘optimal’ sizing of the aortic annulus are mentioned below (Fig. 15.3), however, reader should look into expert consensus documents/recommendations for further in-depth information [15, 40]. The first step is to lock the orthogonal planes at 90° to each other. The cross hairs in the coronal and sagittal images

15 Transcatheter Aortic Valve Implantation (TAVI)

261

a

b

c

d

e

f

g

h

i

Fig. 15.3 Steps to achieve optimal sizing of aortic annulus. Step 1: Volumetric data is loaded onto the workstation and in a standard 3D format (panel 1), the views are transverse (a), Coronal (b) and sagittal (c). The first step is to lock the orthogonal planes at 90° to each other (a–c). Step 2: The cross hairs in the coronal and sagittal images are moved to the level of aortic valves and adjusted so that the cross hairs on the transverse plane are positioned at the level of aortic valve (panel 2 – d–f). Step 3: Then the cross hairs on the coronal and sagittal views are rotated and adjusted so as to correspond to the aortic annulus plane (panel 3). Now the corresponding views are oblique set of images, for e.g. transverse/coronal/ sagittal oblique (g–i). Step 4: The cross hairs on the transverse oblique are rotated either clockwise or anticlockwise (panel 3 g) so as to check whether the cross hairs correspond to aortic annular plane i.e. nadir point of the coronary cusps (panel 4 – j–l). In this panel views, the cross hairs cuts across the cusp rather than located at the hinge/nadir point (panel 4 – k). Step 5: The cross hairs are further adjusted by placing the cross hairs at the aortic annular plane (panel 5 – m–o). Once again the cross hair on the transverse oblique view is rotated and again investigated to see the cross

hairs on the rest of the two oblique planes correspond to aortic annulus plane (panel 6 – p–r). Any adjustments, if necessary, are made by following the above steps. Step 6: Once a satisfactory plane is achieved (panel 7 – s–u), it is further confirmed by toggling the transverse oblique set of images cranially and caudally across the aortic root (panel 7 – s and panel 8 – v, red arrows). If a true aortic annular plane has been achieved, then one would observe that the valve hinge points would disappear or appear symmetrically all at once (panels 7 – s–u and 8 – v–x). Utmost focus is given so that cross hairs are just touching the nadir points (panel 7 – s–u). If this plane is not achieved, then the coronal (panel 7 – t) and sagittal oblique (panel 7 – u) planes needs to be further fine-tuned so as to achieve desired plane and again cross checked by rotating the cross hairs on the transverse oblique view (panel 7 – s). Step 7: Once an optimal view is achieved then the cross hairs are at the true aortic annular plane (panel 9 – y, z, A). Then a ROI trace is made along the circumference of the aortic annulus to calculate the major and minor diameters, mean diameters, area and circumference of the aortic annulus (panel 9 – y). Corresponding coronal and sagittal oblique planes are shown (panel 9 – z, a)

262

C.D. Arepalli et al.

j

k

l

m

n

o

p

q

r

Fig. 15.3 (continued)

15 Transcatheter Aortic Valve Implantation (TAVI)

263

s

t

u

v

w

x

y

z

Fig. 15.3 (continued)

should be moved to the level of aortic valves and adjusted so that the transverse plane is positioned at the level of aortic valve. This would correspond to coronal and sagittal oblique planes respectively. Next the transverse oblique plane is moved up and down across the aortic root so as to visualize the valve hinge points. The valve hinge points should disappear or appear symmetrically all at once, if not, coronal and sagittal oblique planes need to be further fine-tuned so as to achieve desired plane. Again, the cross hairs are rotated

across the transverse oblique so that to visualize the nadir point of the valves in coronal oblique and sagittal oblique planes. Utmost focus is given so that cross hairs are just touching the nadir points and they disappear equally in all planes. Once this is confirmed, the specific position of aortic annulus in the transverse oblique plane is identified and a ROI trace is made along the circumference of the aortic annulus. Majority of the present day work stations have automatic tools that would calculate diameters (D) – Dmin,

264

C.D. Arepalli et al.

Dmax, Dmean, area and perimeter once the tracing is made. If these features are not available, diameter (D) can be calculated from circumference (D = circumference / π). (π = Pi). Area is calculated from π.D2/4.

angiographic projections and MDCT of aortic root [44]. It is important to recognize that correlation with the fluoroscopic angles in the hybrid operating room relies on comparable positioning at the time of CT and during the procedure.

Implantation View – CO Planar Angle Prediction

Evaluation of Aortic Annulus and Aortic Valve Leaflets

The Aortic annulus is not orthogonal to the body axis/planes. For a successful implantation, the prosthetic device needs to be deployed coaxially to the centerline of the aorta. Inappropriate deployment might lead to complications [41]. This plane can be achieved on both CT workstation and fluoroscopy units. However to achieve the same plane in fluoroscopy unit, multiple aortograms are to be performed in a stepwise approach. This situation offers an opportunity for pre-procedural CT to help predict the co-planar angles of deployment prior to the procedure. At here are innumerable co-planar angles of deployment typically ranging from RAO Caudal to LAO Cranial. Preprocedural co-planar angle guidance has been shown to help reduce procedure fluoroscopy time and the volume of contrast needed for the procedure [42, 43]. Binder et al. have demonstrated significant correlation (r = 0.682, p < 0.001) between the predicted optimal deployment projections using 3D

Apart from annulus sizing, assessment of the aortic annulus, left ventricular outflow tract and aortic valve leaflets is equally important for a successful TAVI procedure. Valve calcifications may be distributed focally or diffusely over the surface of the valve cusps or they may present along the leaflet edges. Presence of severe calcium is likely to cause important peri and post procedural complications. Post dilatation techniques might need to be employed in patients with severe calcification. In TAVI cohorts, it is not uncommon to see sub annular or LVOT calcification (Figs. 15.4 and 15.5) or aorto-mitral calcification as a continuum (Fig. 15.4a, b). Distribution of calcium in each of these locations has been shown to have important prognostic implications. In a study by Barbanti et al., 31 consecutive TAVI patients who had LVOT/annular/aortic contained/noncontained rupture were caliper-matched to control group of 31 patients without annular injury [45]. They observed at least 2 features were significantly associated with

a

Fig. 15.4 (a) Left ventricular out flow tract (LVOT) (green arrow) and (b) Mitral annular calcification (MAC) (yellow chevron). Severe LVOT calcification is associated with increased annular rupture. The impact of MAC on TAVI procedure and outcomes is unclear. Presence of MAC might cause stenosis and/or regurgitation and that influences LV

b

function and geometry. It should be noted that deployment of a balloon expandable prosthesis is contraindicated in concomitant significant mitral stenosis with significant mitral annular calcification. LVEF left ventricle ejection fraction

15 Transcatheter Aortic Valve Implantation (TAVI)

a

265

b

Fig. 15.5 Factors influencing annular rupture. (a) Subannular (yellow chevrons) and (b) LVOT calcification (green arrow). Note the relatively low left coronary ostial-aortic annulus height (measured 9.6 mm) (red double arrow)

annular rupture: 1. Moderate/severe LVOT/subannular calcification ((odds ratio, 10.92; 95 % CI, 3.23–36.91; P < 0.001) and 2. ≥20 % area oversizing (odds ratio, 8.38; 95 % CI, 2.67– 26.33; P = 4 mm, adherent annular calcification >6 mm predicted significant PAR (p = 0.03 and 0.009 respectively). Further, it was observed total left and non-coronary and left coronary calcium size score predicted relevant leaks (p = 0.004 and p = 0.001, resp.) but not right or non-coronary cusps calcium [46]. During the procedure, severe calcification may offer resistance to the expansion caused by balloon or selfexpandable devices and may prevent appropriate alignment and/or complete apposition of the sealing skirt to the native commissures. Further, incomplete apposition of the device with commissures may potential lead to development of paravalvular regurgitation (PAR). It has been shown that even mild paravalvular leaks are associated with increased mortality [47, 48] Aortic valve calcifications (AVC) distribution is better visualized on the cross-sectional view of the sinus of Valsalva. AVC can be assessed qualitatively as mild, moderate or severe

(Fig. 15.1a–c) or could also be quantified analogous to the concept of the Agatston score that is being used in assessment of coronary artery calcifications. Haensig et al. assessed for association between native AVC and paravalvular leak in 120 consecutive patients who proceeded with Edwards SAPIEN prosthesis. No paravalvular leaks (n = 66) were noted with mean AVC score of 2704 ± 1510, mild paravalvular leaks (n = 31) with 3804 ± 2739 (P = 0.05); and moderate paravalvular leaks (n = 4) with AVC score of 7387 ± 1044 (P = 0.002) [48]. A significant association between the AVC score and paravalvular leaks [odds ratio [(OR; per AVC score of 1000), 11.38; 95 % confidence interval (CI) 2.33–55.53; P = 0.001)] was noted. It is not only the amount of calcification but location/distribution at each separate cusp or commissure has been found to be associated with PAR [48]. Their study identified significant association for the right and left coronary cusp, for right-left and left non coronary commissure, and there was no significant association for non-coronary right commissure and for non-coronary cusp. Interestingly a number of other studies evaluating patients who underwent TAVI using a self-expanding system,, did not find an association between PAR and the degree of calcification [49, 50]. It is likely the distribution of radial force at different locations due to the nature of the device with balloon expandable at the annulus whereas the self-expandable at the ascending aorta cause varied propensity towards development of PAR.

266

a

C.D. Arepalli et al.

b

Fig. 15.6 Coronary ostial-aortic annulus heights. (a) Low left coronary ostial height (~8 mm) and (b) adequate right coronary ostial height (~13 mm)

Coronary Arteries and Ostial Heights In general, the right and left coronary arteries arise from their respective right and left coronary cusps. They usually arise below the sinotubular junction with the right coronary artery typically at a higher level than left. Given the nature of TAVI, where in the native aortic leaflets are displaced and the bioprosthesis is implanted in the aortic root; the coronary ostial height clearance measurement is important to understand and discriminate those at risk of coronary arterial occlusion. Coronary ostial height measurements are performed in a perpendicular fashion from the annular plane to the coronary ostia (Fig. 15.6a, b). A coronary ostial cut off of 10 mm was suggested by the American College of Cardiology Foundation/ American Association for Thoracic Surgery/Society for Cardiovascular Angiography and Interventions/Society of Thoracic Surgeons expert consensus [15, 51]. However, in a study of reported cases of coronary obstruction, the mean height was 10.3 mm (range 7–12 mm) and 60 % of cases with obstruction had a height >10 mm [52]. In a multicenter registry study, it was observed majority of the patients who had coronary obstruction (~80 % overall and 96 % of women) had a left ostial height 12 mm, 21.4 % had a coronary obstruction implying that there are factors beyond ‘safety cutoff’ of coronary ostial height that might

predispose to coronary obstruction. This registry allowed the field to understand mechanisms of coronary occlusion beyond coronary height that the majority of patients (~ 65 %) who had coronary obstruction had aortic root effacement and also SOV diameter of 80 %) likely related to the female anatomy with smaller aortic root/ shallow aortic sinus and also relatively due to low lying coronary ostia (Figs. 15.5b and 15.6a). In addition, a significant difference in SOV/AA ratios was also observed between those who had coronary obstruction when compared to control groups (1.25 + _0.04 vs. 1.34 + _0.03; p = 0.003) [53]. It is also interesting to note that coronary obstruction was more often observed in patients with balloon expandable devices than those who had self-expandable devices [53, 54]. It is unknown whether makeup of the device or mechanism of deployment has any role in the observed obstructions. Other causes which predispose to coronary obstruction include bulky calcification of the native aortic leaflets, long mobile leaflets and specifically presence of leaflet length more than the coronary ostial heights. Although, at present there are no standard guidelines for a ‘safe’ coronary ostial height; various factors that predispose to coronary obstruction need to be taken into account. These include patient’s sex, the size of the aortic root, sinus of Valsalva diameter, coronary ostial heights, and the ratio of SOV/AA (Fig. 15.7).

15 Transcatheter Aortic Valve Implantation (TAVI)

267

a

b

c

d

Fig. 15.7 Unfavorable anatomy of aortic root (same patient as in Fig. 15.5). (a, b) Aortic annulus is smaller in size and it shows subannular and LVOT calcification (yellow chevron’s). (Even though aortic valvular calcification is of mild degree (c), the aortic sinus is shallow as it measures 18 F), accurate assessment of the vascular access and pathway is critical. As most TAVI cohorts are elderly, it is not uncommon to see vessels which are tortuous and have considerable atherosclerotic burden. In a study of 100 TAVI patients for peripheral artery disease, one-third (35 %) had at least one criterion of unsuitable iliofemoral anatomy and out of those more than 75 % had a luminal diameter of less than 8 mm [62]. The other unfavorable factors included severe circumferential calcification at the iliac bifurcation (>60 %), and severe angulation of the iliac arteries ( 1.05 and SFAAR > 1.35 were associated with increased vascular complications. SFAR Sheath to Femoral Artery Ratio (Diameter), SFAAR Sheath Area to Femoral Minimal Lumen Area Ratio

270

C.D. Arepalli et al.

of circumferential ilio-femoral calcifications was an important risk factor for vascular complications and also an independent predictor of increased mortality after TF-TAVI [65]. Further, incorporation of this MSCT derived parameter in the workup algorithm of patients with a sheath-to-iliofemoral artery ratio – SIFAR ≥1 on angiographic screening improved the specificity for prediction of major vascular complications to 62 % without altering the sensitivity (100 %). In another series, arterial calcification (Grade > 2) at puncture site was found to be independent predictor of major vascular complication, p < 0.001 [66] . The authors also noted despite the use of a pigtail catheter for an ideal puncture, arterial wall calcification could not be evaluated by fluoroscopy and recommended assessment through MDCT. Kurra et al. proposed presence of more than 60 % circumferential calcification at the external–internal iliac bifurcation as an unsuitable iliofemoral anatomy [62].

access vessel diameter (TT/AD) index and observed that a value 27.89 predicts vascular complications with 50.8 % sensitivity and 70.6 % specificity (AUC: 0.22, P = 0.008) [66]. TT/AD ratio defined as sum of angles (TT) divided by the minimum femoral arterial diameter (AD) at the access site. Even though, the indexed values were primarily based on 2D invasive angiography, nevertheless, a small subset of their cohort were evaluated with 3D MDCT and found to have good correlation (r = 0.66, p = 0.013). Chiam et al. assessed iliofemoral dimensions and characteristics by ultrasonography in 549 Asian patients [68]. They observed female gender, lower body surface area, and presence of diabetes mellitus; dyslipidemia and smoking history were independent factors for smaller iliofemoral dimensions. Thus for the same degree of tortuosity, a smaller intraluminal diameter might predispose to increased vascular risk, if TT/ AD index is incorporated.

Iliofemoral Arterial Tortuosity

Alternative Access

Iliofemoral Arterial Tortuosity have implications for transfemoral TAVI approach as excess tortuosity might increase the access site complication rates and thus would affect procedural success. The degree of tortuosity could be graded as – 0-no tortuosity; 1-mild tortuosity (30–60°); 2-moderate tortuosity (60–90°); and 3-severe tortuosity (≥90°) (Fig. 15.10) [62, 64]. Although it was observed that iliofemoral tortuosity alone does not predict vascular complications [58, 67]; Vavuranakis et al. in their study incorporated total arterial tortuosity/

Alternative access routes should be considered in patients where reconstructed CT images reveal unfavorable ilio-femoral anatomy. Description of various access sites is beyond the scope of this chapter; however, the underlying principles would remain the same as mentioned above. Access routes need to be assessed holistically depending upon patient’s condition and also the device that is likely to be deployed with minimal complications (Fig. 15.11).

a

b

Fig. 15.10 Ilio-femoral tortuosity. Multiple panels show mild, moderate and severe tortuosity of peripheral vascular access (a–c). Degree of tortuosity is graded as: 0-no tortuosity; (a) 1-mild tortuosity (30–60°); (b) 2-moderate tortuosity (60–90°); and (c)

c

3-severe tortuosity (≥90°) [64, 66]. Although, the degree of tortuosity per se might not significantly influence increased risk for vascular complications; a combination of increased tortuosity and small luminal diameter predisposes to vascular risk

15 Transcatheter Aortic Valve Implantation (TAVI)

a

271

b

c

Fig. 15.11 Structural overlay of peripheral access sites. MDCT volumetric acquisition provides VRT images which help in better anatomical understanding of the overlying structures. For e.g. (a). Overlay of Iliofemoral arteries against the backdrop of pelvic bones helps in easy assessment of tortuosity. In another TAVI patient, an unfavorable iliofemoral access was noted and hence alternate

Conclusion

The integration of computed tomography into transcatheter aortic valve replacement planning and guidance has seen rapid progression over the last 5 years. CT has gone

vascular access sites were assessed (b, c). VRT images provide visual depiction of prior bypass graft procedure and presence of left brachiocephalic vein overlying the ascending aorta (b, c). If a transaortic access is contemplated relevant knowledge of anatomy and their relations to each other help in proper planning. VRT volume rendered technique

from simply being a tool for the assessment of iliofemoral access to now being the non-invasive test of choice for preprocedural annular sizing and transcatheter heart valve selection. With ongoing evolution of the transcatheter

272

devices and the procedure, as well as the introduction of other transcatheter valvular solutions CT will almost certainly grow in its utilization and help us further understand how to appropriately size transcatheter devices and hopefully reduce procedural related complications.

References 1. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Borden WB, et al. Heart disease and stroke statistics – 2013 update: a report from the American Heart Association. Circulation. 2013;127(1):e6– 245. Epub 2012/12/15. 2. Thaden JJ, Nkomo VT, Enriquez-Sarano M. The global burden of aortic stenosis. Prog Cardiovasc Dis. 2014;56(6):565–71. Epub 2014/05/20. 3. Osnabrugge RL, Mylotte D, Head SJ, Van Mieghem NM, Nkomo VT, LeReun CM, et al. Aortic stenosis in the elderly: disease prevalence and number of candidates for transcatheter aortic valve replacement: a meta-analysis and modeling study. J Am Coll Cardiol. 2013;62(11):1002–12. Epub 2013/06/04. 4. Nkomo VT, Gardin JM, Skelton TN, Gottdiener JS, Scott CG, Enriquez-Sarano M. Burden of valvular heart diseases: a population-based study. Lancet. 2006;368(9540):1005–11. Epub 2006/09/19. 5. Fedak PW, Verma S, David TE, Leask RL, Weisel RD, Butany J. Clinical and pathophysiological implications of a bicuspid aortic valve. Circulation. 2002;106(8):900–4. Epub 2002/08/21. 6. Beppu S, Suzuki S, Matsuda H, Ohmori F, Nagata S, Miyatake K. Rapidity of progression of aortic stenosis in patients with congenital bicuspid aortic valves. Am J Cardiol. 1993;71(4):322–7. Epub 1993/02/01. 7. Pomerance A. Pathogenesis of aortic stenosis and its relation to age. Br Heart J. 1972;34(6):569–74. Epub 1972/06/01. 8. Mautner GC, Mautner SL, Cannon 3rd RO, Hunsberger SA, Roberts WC. Clinical factors useful in predicting aortic valve structure in patients >40 years of age with isolated valvular aortic stenosis. Am J Cardiol. 1993;72(2):194–8. Epub 1993/07/15. 9. Ross Jr J, Braunwald E. Aortic stenosis. Circulation. 1968;38(1 Suppl):61–7. Epub 1968/07/01. 10. Salinas P, Moreno R, Lopez-Sendon JL. Transcatheter aortic valve implantation: current status and future perspectives. World J Cardiol. 2011;3(6):177–85. Epub 2011/07/21. 11. Panchal HB, Ladia V, Desai S, Shah T, Ramu V. A meta-analysis of mortality and major adverse cardiovascular and cerebrovascular events following transcatheter aortic valve implantation versus surgical aortic valve replacement for severe aortic stenosis. Am J Cardiol. 2013;112(6):850–60. Epub 2013/06/13. 12. Cao C, Ang SC, Indraratna P, Manganas C, Bannon P, Black D, et al. Systematic review and meta-analysis of transcatheter aortic valve implantation versus surgical aortic valve replacement for severe aortic stenosis. Ann Cardiothorac Surg. 2013;2(1):10–23. Epub 2013/08/27. 13. Cribier A, Eltchaninoff H, Bash A, Borenstein N, Tron C, Bauer F, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation. 2002;106(24):3006–8. Epub 2002/12/11. 14. Leipsic J, Yang TH, Min JK. Computed tomographic imaging of transcatheter aortic valve replacement for prediction and prevention of procedural complications. Circ Cardiovasc Imaging. 2013;6(4): 597–605. Epub 2013/07/19.

C.D. Arepalli et al. 15. Achenbach S, Delgado V, Hausleiter J, Schoenhagen P, Min JK, Leipsic JA. SCCT expert consensus document on computed tomography imaging before transcatheter aortic valve implantation (TAVI)/transcatheter aortic valve replacement (TAVR). J Cardiovasc Comput Tomogr. 2012;6(6):366–80. Epub 2012/12/12. 16. Neragi-Miandoab S, Michler RE. A review of most relevant complications of transcatheter aortic valve implantation. ISRN Cardiol. 2013;2013:956252. Epub 2013/07/12. 17. Tops LF, Wood DA, Delgado V, Schuijf JD, Mayo JR, Pasupati S, et al. Noninvasive evaluation of the aortic root with multislice computed tomography implications for transcatheter aortic valve replacement. JACC Cardiovasc Imaging. 2008;1(3):321–30. Epub 2009/04/10. 18. Schultz CJ, Moelker A, Piazza N, Tzikas A, Otten A, Nuis RJ, et al. Three dimensional evaluation of the aortic annulus using multislice computer tomography: are manufacturer’s guidelines for sizing for percutaneous aortic valve replacement helpful? Eur Heart J. 2010;31(7):849–56. Epub 2009/12/10. 19. Blanke P, Russe M, Leipsic J, Reinohl J, Ebersberger U, Suranyi P, et al. Conformational pulsatile changes of the aortic annulus: impact on prosthesis sizing by computed tomography for transcatheter aortic valve replacement. JACC Cardiovasc Interv. 2012;5(9): 984–94. Epub 2012/09/22. 20. Ng AC, Yiu KH, Ewe SH, van der Kley F, Bertini M, de Weger A, et al. Influence of left ventricular geometry and function on aortic annular dimensions as assessed with multi-detector row computed tomography: implications for transcatheter aortic valve implantation. Eur Heart J. 2011;32(22):2806–13. Epub 2011/07/26. 21. Nakai H, Takeuchi M, Yoshitani H, Kaku K, Haruki N, Otsuji Y. Pitfalls of anatomical aortic valve area measurements using twodimensional transoesophageal echocardiography and the potential of three-dimensional transoesophageal echocardiography. Eur J Echocardiogr. 2010;11(4):369–76. Epub 2009/12/22. 22. Bertaso AG, Wong DT, Liew GY, Cunnington MS, Richardson JD, Thomson VS, et al. Aortic annulus dimension assessment by computed tomography for transcatheter aortic valve implantation: differences between systole and diastole. Int J Cardiovasc Imaging. 2012;28(8):2091–8. Epub 2012/02/10. 23. Ng AC, Delgado V, van der Kley F, Shanks M, van de Veire NR, Bertini M, et al. Comparison of aortic root dimensions and geometries before and after transcatheter aortic valve implantation by 2- and 3-dimensional transesophageal echocardiography and multislice computed tomography. Circ Cardiovasc Imaging. 2010; 3(1):94–102. Epub 2009/11/19. 24. Jilaihawi H, Kashif M, Fontana G, Furugen A, Shiota T, Friede G, et al. Cross-sectional computed tomographic assessment improves accuracy of aortic annular sizing for transcatheter aortic valve replacement and reduces the incidence of paravalvular aortic regurgitation. J Am Coll Cardiol. 2012;59(14):1275–86. Epub 2012/03/01. 25. Hamdan A, Guetta V, Konen E, Goitein O, Segev A, Raanani E, et al. Deformation dynamics and mechanical properties of the aortic annulus by 4-dimensional computed tomography: insights into the functional anatomy of the aortic valve complex and implications for transcatheter aortic valve therapy. J Am Coll Cardiol. 2012;59(2):119–27. Epub 2012/01/10. 26. Buellesfeld L, Stortecky S, Kalesan B, Gloekler S, Khattab AA, Nietlispach F, et al. Aortic root dimensions among patients with severe aortic stenosis undergoing transcatheter aortic valve replacement. JACC Cardiovasc Interv. 2013;6(1):72–83. Epub 2013/01/26. 27. Zhang R, Song Y, Zhou Y, Sun L. Comparison of aortic annulus diameter measurement between multi-detector computed tomography and echocardiography: a meta-analysis. PLoS One. 2013; 8(3):e58729. Epub 2013/03/22.

15 Transcatheter Aortic Valve Implantation (TAVI) 28. Mizia-Stec K, Pysz P, Jasinski M, Adamczyk T, DrzewieckaGerber A, Chmiel A, et al. Preoperative quantification of aortic valve stenosis: comparison of 64-slice computed tomography with transesophageal and transthoracic echocardiography and size of implanted prosthesis. Int J Cardiovasc Imaging. 2012;28(2):343– 52. Epub 2011/02/01. 29. Messika-Zeitoun D, Serfaty JM, Brochet E, Ducrocq G, Lepage L, Detaint D, et al. Multimodal assessment of the aortic annulus diameter: implications for transcatheter aortic valve implantation. J Am Coll Cardiol. 2010;55(3):186–94. Epub 2010/02/02. 30. Koos R, Altiok E, Mahnken AH, Neizel M, Dohmen G, Marx N, et al. Evaluation of aortic root for definition of prosthesis size by magnetic resonance imaging and cardiac computed tomography: implications for transcatheter aortic valve implantation. Int J Cardiol. 2012;158(3):353–8. Epub 2011/02/15. 31. Willson AB, Webb JG, Labounty TM, Achenbach S, Moss R, Wheeler M, et al. 3-dimensional aortic annular assessment by multidetector computed tomography predicts moderate or severe paravalvular regurgitation after transcatheter aortic valve replacement: a multicenter retrospective analysis. J Am Coll Cardiol. 2012;59(14):1287–94. Epub 2012/03/01. 32. Altiok E, Koos R, Schroder J, Brehmer K, Hamada S, Becker M, et al. Comparison of two-dimensional and three-dimensional imaging techniques for measurement of aortic annulus diameters before transcatheter aortic valve implantation. Heart (Br Cardiac Soc). 2011;97(19):1578–84. Epub 2011/06/28. 33. Husser O, Rauch S, Endemann DH, Resch M, Nunez J, Bodi V, et al. Impact of three-dimensional transesophageal echocardiography on prosthesis sizing for transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2012;80(6):956–63. Epub 2012/03/16. 34. Binder RK, Webb JG, Willson AB, Urena M, Hansson NC, Norgaard BL, et al. The impact of integration of a multidetector computed tomography annulus area sizing algorithm on outcomes of transcatheter aortic valve replacement: a prospective, multicenter, controlled trial. J Am Coll Cardiol. 2013;62(5):431–8. Epub 2013/05/21. 35. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, et al. Transcatheter aortic-valve replacement with a selfexpanding prosthesis. N Engl J Med. 2014;370(19):1790–8. Epub 2014/04/01. 36. Lehmkuhl L, Foldyna B, Von Aspern K, Lucke C, Grothoff M, Nitzsche S, et al. Inter-individual variance and cardiac cycle dependency of aortic root dimensions and shape as assessed by ECG-gated multi-slice computed tomography in patients with severe aortic stenosis prior to transcatheter aortic valve implantation: is it crucial for correct sizing? Int J Cardiovasc Imaging. 2013;29(3):693–703. Epub 2012/09/19. 37. de Heer LM, Budde RP, van Prehn J, Mali WP, Bartels LW, Stella PR, et al. Pulsatile distention of the nondiseased and stenotic aortic valve annulus: analysis with electrocardiogram-gated computed tomography. Ann Thorac Surg. 2012;93(2):516–22. Epub 2011/12/06. 38. von Aspern K, Foldyna B, Etz CD, Hoyer A, Girrbach F, Holzhey D, et al. Effective diameter of the aortic annulus prior to transcatheter aortic valve implantation: influence of area-based versus perimeter-based calculation. Int J Cardiovasc Imaging. 2015;31: 163–9. Epub 2014/08/29. 39. Blanke P, Willson AB, Webb JG, Achenbach S, Piazza N, Min JK, et al. Oversizing in transcatheter aortic valve replacement, a commonly used term but a poorly understood one: dependency on definition and geometrical measurements. J Cardiovasc Comput Tomogr. 2014;8(1):67–76. Epub 2014/03/04. 40. Kasel AM, Cassese S, Bleiziffer S, Amaki M, Hahn RT, Kastrati A, et al. Standardized imaging for aortic annular sizing: implications for transcatheter valve selection. JACC Cardiovasc Imaging. 2013;6(2):249–62. Epub 2013/03/16.

273 41. Kurra V, Kapadia SR, Tuzcu EM, Halliburton SS, Svensson L, Roselli EE, et al. Pre-procedural imaging of aortic root orientation and dimensions: comparison between X-ray angiographic planar imaging and 3-dimensional multidetector row computed tomography. JACC Cardiovasc Interv. 2010;3(1):105–13. Epub 2010/02/05. 42. Gurvitch R, Wood DA, Leipsic J, Tay E, Johnson M, Ye J, et al. Multislice computed tomography for prediction of optimal angiographic deployment projections during transcatheter aortic valve implantation. JACC Cardiovasc Interv. 2010;3(11):1157–65. Epub 2010/11/23. 43. Leipsic J, Gurvitch R, Labounty TM, Min JK, Wood D, Johnson M, et al. Multidetector computed tomography in transcatheter aortic valve implantation. JACC Cardiovasc Imaging. 2011;4(4):416–29. Epub 2011/04/16. 44. Binder RK, Leipsic J, Wood D, Moore T, Toggweiler S, Willson A, et al. Prediction of optimal deployment projection for transcatheter aortic valve replacement: angiographic 3-dimensional reconstruction of the aortic root versus multidetector computed tomography. Circ Cardiovasc Interv. 2012;5(2):247–52. Epub 2012/03/23. 45. Barbanti M, Yang TH, Rodes Cabau J, Tamburino C, Wood DA, Jilaihawi H, et al. Anatomical and procedural features associated with aortic root rupture during balloon-expandable transcatheter aortic valve replacement. Circulation. 2013;128(3):244–53. Epub 2013/06/12. 46. Feuchtner G, Plank F, Bartel T, Bonaros N, Müller S, Leipsic J, et al. TCT-836 prediction of paravalvular leaks after transcatheter aortic valve implantation by valvular or annular calcification? J Am Coll Cardiol. 2012;60(17_S). 47. Kodali SK, Williams MR, Smith CR, Svensson LG, Webb JG, Makkar RR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012;366(18):1686–95. Epub 2012/03/27. 48. Haensig M, Lehmkuhl L, Rastan AJ, Kempfert J, Mukherjee C, Gutberlet M, et al. Aortic valve calcium scoring is a predictor of significant paravalvular aortic insufficiency in transapical-aortic valve implantation. Eur J Cardiothorac Surg. 2012;41(6):1234–40. discussion 40-1. Epub 2012/01/14. 49. John D, Buellesfeld L, Yuecel S, Mueller R, Latsios G, Beucher H, et al. Correlation of Device landing zone calcification and acute procedural success in patients undergoing transcatheter aortic valve implantations with the self-expanding CoreValve prosthesis. JACC Cardiovasc Interv. 2010;3(2):233–43. Epub 2010/02/23. 50. Koos R, Mahnken AH, Dohmen G, Brehmer K, Gunther RW, Autschbach R, et al. Association of aortic valve calcification severity with the degree of aortic regurgitation after transcatheter aortic valve implantation. Int J Cardiol. 2011;150(2):142–5. Epub 2010/03/31. 51. Holmes Jr DR, Mack MJ, Kaul S, Agnihotri A, Alexander KP, Bailey SR, et al. 2012 ACCF/AATS/SCAI/STS expert consensus document on transcatheter aortic valve replacement: developed in collaboration with the American Heart Association, American Society of Echocardiography, European Association for Cardio-Thoracic Surgery, Heart Failure Society of America, Mended Hearts, Society of Cardiovascular Anesthesiologists, Society of Cardiovascular Computed Tomography, and Society for Cardiovascular Magnetic Resonance. J Thorac Cardiovasc Surg. 2012;144(3):e29–84. Epub 2012/08/18. 52. Ribeiro HB, Nombela-Franco L, Urena M, Mok M, Pasian S, Doyle D, et al. Coronary obstruction following transcatheter aortic valve implantation: a systematic review. JACC Cardiovasc Interv. 2013; 6(5):452–61. Epub 2013/04/23. 53. Ribeiro HB, Webb JG, Makkar RR, Cohen MG, Kapadia SR, Kodali S, et al. Predictive factors, management, and clinical outcomes of coronary obstruction following transcatheter aortic valve

274

54.

55.

56.

57.

58.

59.

60.

61.

C.D. Arepalli et al. implantation: insights from a large multicenter registry. J Am Coll Cardiol. 2013;62(17):1552–62. Epub 2013/08/21. Khatri PJ, Webb JG, Rodes-Cabau J, Fremes SE, Ruel M, Lau K, et al. Adverse effects associated with transcatheter aortic valve implantation: a meta-analysis of contemporary studies. Ann Intern Med. 2013;158(1):35–46. Epub 2013/01/02. Genereux P, Webb JG, Svensson LG, Kodali SK, Satler LF, Fearon WF, et al. Vascular complications after transcatheter aortic valve replacement: insights from the PARTNER (Placement of AoRTic TraNscathetER Valve) trial. J Am Coll Cardiol. 2012;60(12):1043– 52. Epub 2012/08/14. Genereux P, Head SJ, Van Mieghem NM, Kodali S, Kirtane AJ, Xu K, et al. Clinical outcomes after transcatheter aortic valve replacement using valve academic research consortium definitions: a weighted meta-analysis of 3,519 patients from 16 studies. J Am Coll Cardiol. 2012;59(25):2317–26. Epub 2012/04/17. Mwipatayi BP, Picardo A, Masilonyane-Jones TV, Larbalestier R, Thomas S, Turner J, et al. Incidence and prognosis of vascular complications after transcatheter aortic valve implantation. J Vasc Surg. 2013;58(4):1028–36.e1. Epub 2013/09/03. Hayashida K, Lefevre T, Chevalier B, Hovasse T, Romano M, Garot P, et al. Transfemoral aortic valve implantation new criteria to predict vascular complications. JACC Cardiovasc Interv. 2011; 4(8):851–8. Epub 2011/08/20. Lange R, Bleiziffer S, Piazza N, Mazzitelli D, Hutter A, TassaniPrell P, et al. Incidence and treatment of procedural cardiovascular complications associated with trans-arterial and trans-apical interventional aortic valve implantation in 412 consecutive patients. Eur J Cardiothorac Surg. 2011;40(5):1105–13. Epub 2011/04/26. Leon MB, Piazza N, Nikolsky E, Blackstone EH, Cutlip DE, Kappetein AP, et al. Standardized endpoint definitions for Transcatheter Aortic Valve Implantation clinical trials: a consensus report from the Valve Academic Research Consortium. J Am Coll Cardiol. 2011;57(3):253–69. Epub 2011/01/11. Kappetein AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, et al. Updated standardized endpoint definitions

62.

63.

64.

65.

66.

67.

68.

for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. J Am Coll Cardiol. 2012;60(15):1438–54. Epub 2012/10/06. Kurra V, Schoenhagen P, Roselli EE, Kapadia SR, Tuzcu EM, Greenberg R, et al. Prevalence of significant peripheral artery disease in patients evaluated for percutaneous aortic valve insertion: preprocedural assessment with multidetector computed tomography. J Thorac Cardiovasc Surg. 2009;137(5):1258–64. Epub 2009/04/22. Krishnaswamy A, Parashar A, Agarwal S, Modi DK, Poddar KL, Svensson LG, et al. Predicting vascular complications during transfemoral transcatheter aortic valve replacement using computed tomography: a novel area-based index. Catheter Cardiovasc Interv. 2014;84(5):844–51. Epub 2014/03/25. Eltchaninoff H, Kerkeni M, Zajarias A, Tron C, Godin M, Sanchez Giron C, et al. Aorto-iliac angiography as a screening tool in selecting patients for transfemoral aortic valve implantation with the Edwards SAPIEN bioprosthesis. EuroIntervention. 2009;5(4):438– 42. Epub 2009/09/17. Reinthaler M, Aggarwal SK, De Palma R, Landmesser U, Froehlich G, Yap J, et al. Predictors of clinical outcome in transfemoral TAVI: Circumferential iliofemoral calcifications and manufacturerderived recommendations. Anadolu kardiyoloji dergisi: AKD Anatol J Cardiol. 2015;15:297–305. Epub 2014/11/22. Vavuranakis M, Kariori M, Voudris V, Kalogeras K, Vrachatis D, Aznaouridis C, et al. Predictive factors of vascular complications after transcatheter aortic valve implantation in patients treated with a default percutaneous strategy. Cardiovasc Ther. 2013;31(5):e46– 54. Epub 2013/06/15. Toggweiler S, Gurvitch R, Leipsic J, Wood DA, Willson AB, Binder RK, et al. Percutaneous aortic valve replacement: vascular outcomes with a fully percutaneous procedure. J Am Coll Cardiol. 2012;59(2):113–8. Epub 2012/01/10. Chiam PT, Koh AS, Ewe SH, Sin YK, Chao VT, Ng CK, et al. Iliofemoral anatomy among Asians: implications for transcatheter aortic valve implantation. Int J Cardiol. 2013;167(4):1373–9. Epub 2012/04/24.

Assessment of Cardiac and Thoracic Masses

16

Jabi E. Shriki, Patrick M. Colletti, and Suresh Maximin

Abstract

Cardiac masses are uncommonly encountered, but can pose a perplexing diagnostic dilemma when present. Familiarity with cross-sectional imaging of the heart can provide a number of tools to enable diagnosis. In this chapter, we discuss the varied appearances of cardiac masses. Thrombi tend to occur in characteristic locations include the left atrial appendage and left ventricular apex. Benign neoplastic masses, tend to be pedunculated and have narrow attachments to the myocardial walls. For example, atrial myxomas tend to have narrow attachments at the left atrial side of the interatrial septum. Malignant neoplastic masses tend to grow in an infiltrative pattern with broad attachments to the myocardium. Keywords

Myocardial masses • Cardiac mass • Thrombus • Myxoma • Angiosarcoma • Cardiac metastases

Overview Excellent spatial and contrast resolution make cardiovascular computed tomographic angiography (CCTA) an ideal method for the detection and evaluation of cardiac masses and masses adjacent to the heart. Suspended respiration and cardiac gating techniques employed with CCTA enhance delineation of planes between masses and normal structures. While many cardiac masses are well demonstrated with nongated CT, the ability to freeze cardiac motion enables clearer evaluation of tissue attenuation and enhancement character-

J.E. Shriki, MD (*) Department of Radiology, Puget VA Health System, University of Washington, 1660 S. Columbian Way, Seattle, WA 98101, USA e-mail: [email protected] P.M. Colletti, MD Department of Radiology, University of Southern California, Los Angeles, CA, USA S. Maximin, MD Department of Radiology, University of Washington, Seattle, WA, USA

istics of normal myocardium and of masses. Two approaches to cardiac gating may be applied: prospective ECG triggering and retrospective ECG gating [1]. Pre-contrast CT imaging may help to identify some features, such as calcifications, which appear as foci of punctuate or coarse hyperattenuation, typically in the range of 130 Hounsfield units (HU), and hemorrhage, which may have a more modest and ill-defined hyperattenuation relative to normal myocardium. CT without contrast is the imaging modality of choice for demonstrating calcification. Demonstration of low attenuation prior to contrast administration may be helpful The amount of iodinated contrast agent required for satisfactory CT evaluation of cardiac masses depends on patient mass, with 0.5–1.0 g of iodine per kilogram body mass as the usual dose [2]. Most clinically used contrast agents have low osmolality, with concentrations of 300–400 mg iodine/ml. Typically, 100 ml of contrast agent is administered at 4–5 ml/s via a programmable injector system. This is followed by a bolus flush of 50 ml of normal saline [3]. Timing CT image acquisition to the arrival of the contrast bolus in the left atrium or left ventricle is typically

© Springer International Publishing 2016 M.J. Budoff, J.S. Shinbane (eds.), Cardiac CT Imaging: Diagnosis of Cardiovascular Disease, DOI 10.1007/978-3-319-28219-0_16

275

J.E. Shriki et al.

276

identical to the timing used for coronary artery CT examinations [4]. Optimal left atrial appendage enhancement may, however, be somewhat difficult, since the left atrial appendage may opacify somewhat more slowly or heterogeneously. In addition, there may be considerable variability in circulation time from patient to patient, particularly in patients with cardiac tumors or thrombi. The time between intravenous contrast injection and appearance of contrast in the aorta can be determined using a small volume test contrast agent bolus of 20 ml and rapid, repeated imaging of a single trans-aortic plane [5]. Alternatively, with bolus tracking, a Hounsfield unit (HU) threshold may be set such that the volume acquisition is triggered to begin once a certain HU value is detected in the ascending aorta. A uniform, programmed injection requires 10–25 s for delivery of intravenous contrast agent and up to 50 additional seconds for the saline flush. One potential pitfall in employing automatic bolus detection in cardiac masses is that it is possible to inadvertently place the bolus

a

detection region of interest (ROI) within a chamber or a vessel which contains internal thrombus or tumor. Such an error may result in failure to detect the bolus as shown in Fig. 16.1. Opacification of the right heart with contrast may be more challenging. Without appropriate acquisition timing for right heart visualization, there may be insufficient contrast for delineation of right atrial and right ventricular endocardial borders. Excessive contrast within the right heart may result in streaking and obscuration of subtle masses. Even when the right heart is well opacified, there is frequently swirling of contrast with non-opacified blood arriving from the inferior vena cava. The mixture of opacified and non-opacified blood can make delineation of masses in the right heart difficult. Optimal timing and technique for right heart examination usually differs from that used for routine CCTA. In most patients, optimal right ventricular opacification is achieved by placing the ROI for bolus tracking in the main pulmonary artery.

b

H 120 U 100 E n 80 h 60 a n 40 c e 20 m 0 1 e

1

1

1 0

c

1 1

n -20 t

15 5 10 Elapsed Time (sec)

20

Fig. 16.1 (a, b) Pitfall of automated bolus detection. Automatic bolus detection fails due to tumor replacing the blood pool in the selected regionof-interest in the main pulmonary artery

16

Assessment of Cardiac and Thoracic Masses

277

Right ventricular delineation in congenital heart disease with transposition or other abnormal great vessel relationships requires some a priori knowledge of the anatomy and relevant surgical history to select the correct region for timing prescription. When opacification of multiple chambers is needed, more complex injection protocols can be utilized with multiphasic contrast administration, including injection of mixtures of saline and contrast [6]. One advantage cardiac magnetic resonance imaging (CMR) holds over cardiac CT for cardiac masses is the ability to obtain excellent contrast for both the right ventricle and left ventricle in the same examination due to the ability to image the heart in multiple phases of contrast administration, with no radiation dose. Venous or equilibrium phase imaging on CMR can help to homogeneously opacify the right heart chambers. Clinical [7–19] and imaging [20–28] features of cardiac masses are summarized in Tables 16.1, 16.2, 16.3, and 16.4.

Interpreting Cardiac Masses: Key Descriptors Location Lesion location relative to the specific, involved chambers should be noted and may provide a hint as to the nature of a particular mass. Masses related to the heart itself which are cardiac in origin have a unique differential diagnosis. Masses immediately adjacent to the heart and intimately involving the pericardium should be described as pericardial (Fig. 16.2). Masses which are external to the heart should be described as paracardial (within the mediastinum, adjacent to the heart). For lesions within the mediastinum, a separate set of diagnostic possibilities should be included in the differential. A full discussion of mediastinal masses, however, is beyond the scope of this text. Although some tumors may violate planes and make identification of the organ of origin difficult, in most cases, cardiac masses, pericardial masses, and mediastinal masses can be separated.

Table 16.1 Benign cardiac neoplasms Myxoma (40 % of all benign tumors) (Fig. 16.16)

Location LA septum 75 %; RA 18 %; ventricles 7 %; multiple 5 %

Fibroelastoma Lipoma

Arise from valves; project into aorta or MPA Varies

Lipomatous hypertrophy Fibroma

Atrial septum; protrudes into RA Myocardium

Hemangioma Lymphangioma Paraganglioma, dysembryoma, pheochromocytoma

Myocardium Myocardium Paracardiac; AV groove

Teratoma

Pericardial; attach to the aorta or PA roots

Features 10 % calcified; frequent systemic emboli; may protrude through mitral valve during diastole Derived from endocardium; may be multiple; often an incidental finding at surgery Encapsulated adipose tissue (fat attenuation); asymptomatic; negative CT density; 25 % are multiple; consider tuberous sclerosis; should not be confused with fat in paracardiac folds Fat attenuation Well delineated, calcified; enhance minimally Calcifications; delayed enhancement Diffuse proliferation; minimally enhancing Sympathetic plexus; hyper-enhancing; correlate with urinary catecholamines; alpha-and beta-blockade for surgery Multi-cystic; frequently calcify; moderate enhancement

Table 16.2 Cystic cardiac masses Pleuro-pericardial cyst

Location 75 % in right paracardiac angle

Echinococcal cysts

Myocardial or pericardial

Tuberculoma Hematoma

Myocardial or pericardial Posterior recesses at the aortic root or left atrium Course of coronary arteries

Thrombosed coronary aneurysm

Features Asymptomatic (avascular/calcified); unilocular, sharply marginated, 20–40 HU; may communicate with pericardium; change shape with body position (Avascular/calcific rim); nearly always also in liver, lung, eyes, brain Calcified; constrictive pericarditis Acutely hyper-dense; may calcify; traumatic or post-surgical Calcified rim; thrombus

J.E. Shriki et al.

278 Table 16.3 Malignant cardiac tumors Metastasis (20 × as common as primary tumor) (Fig. 16.2)

Lung, breast, melanoma, sarcoma, leukemia, thyroid, kidney Renal, urothelial, hepatocellular, adrenal, retroperitoneal sarcoma Lymphoma

Location Pericardial; intravascular; intra-myocardial

Angiosarcoma (Fig. 16.1)

Pericardial; direct or lymphatic; hematogenous; direct venous extension Extend up the inferior vena cava to the right atrium Pericardium; myocardium; commonly basal in location Pericardium; RV, RA, myocardium

Osteosarcoma Rhabdomyosarcoma, fibrosarcoma

RA, RV Myocardium

Mesothelioma

Pericardium

Features Seen in 10 % of end-stage cancers; lung (36 %), breast (7 %), esophagus (6 %); lymphoma, melanoma, Kaposi’s sarcoma, leukemia (20 %); modes of dissemination: direct or lymphatic; hematogenous; direct venous extension (pulmonary veins or inferior vena cava) Lung cancer may extend to the left atrium along the pulmonary veins Enhancing intravascular mass; primary tumor identified May infiltrate epicardial fat; 50 % associated with HIV Angiosarcoma of the pericardium or right ventricle is most common; poor prognosis; distribution is similar to lymphoma Ossification Most common primary cardiac malignancy in infants and children Always involves the myocardium; pericardial involvement is typically in the form of nodular masses rather than sheet-like spread Intra-pericardial mass; effusions; constrictive physiology

Table 16.4 Other cardiac masses Endo-myocardial fibrosis

Location Pericardium; myocardium

Erdheim-Chester disease

Pericardium; myocardium

RA thrombus (Fig. 16.12) RV thrombus

Right atrium Right ventricle

LA thrombus (Fig. 16.11)

Left atrium

LV thrombus (Fig. 16.8)

Left ventricle

Vegetations

Valves; catheters

Chamber Involvement The chamber of origin and location within the chamber should be noted. For example, a mass in the left atrium attached along the interatrial septum has a higher chance of being an atrial myxoma. A mass in the left atrial appendage has a higher probability of being a thrombus. Some authors have suggested that, on imaging, metastases are more common in the right heart. However, this could be due to the earlier detection of right heart masses, since the wall of the right ventricle is thinner than the wall of the left ventricle. A mass in the left ventricle may be neoplastic, if it is felt to be arising from the

Features Thickened pericardium; thickened myocardium with patchy enhancement restrictive and constrictive physiology Thickened pericardium; thickened (non-langherans fibrosis) myocardium with patchy enhancement restrictive and constrictive physiology Associated with indwelling catheters and devices Associated with severe coagulopathy; dilated cardiomyopathy Seen in atrial fibrillation and mitral stenosis; attached to posterior or superior atrial wall; may be calcified Common complication of myocardial infarction (20–40 % of anterior MIs); contiguous to akinetic myocardium; most common at the apex EKG-triggered cine views of valves helpful

wall. A mass at the apex of the left ventricle, which appears separate from the wall, has a higher probability of being a thrombus. Attention should be given to associated wall motion abnormalities or aneurysms. Severe metastatic involvement of the myocardial wall may result in a wall motion abnormality. However, a thrombus may present as a mass closely associated with a wall motion abnormality such as dyskinetic aneurysmal segment. Transiently, thromboemboli with a peripheral origin, such as deep vein thrombi, may be seen in the right atrium and right ventricle (Fig. 16.3), and are called “in transit thrombi”. A mass which arises from the crista terminalis of the right atrium may be a prominent network of Chiari. Elastofibromas are

16

Assessment of Cardiac and Thoracic Masses

279

a

c

b

d

Fig. 16.2 A 22-year-old female with a primary pericardial primitive neuroectodermal tumor. Images obtained are obtained as part of a postcontrast non-gated CT scan of the chest. Transverse and oblique 4-chamber views are shown (a–d). In this case, the tumor was causing

restriction of cardiac motion, and as a result artifacts due to cardiac motion are mild. There is heavy neoplastic infiltration of the atrioventricular groove with invasion of the right atrium and ventricle (black arrowheads)

common lesions which occur along the valve surfaces, but are usually small and not well-seen on CCTA. Valvular vegetations may rarely grow to a size where they may mimic a cardiac mass, although this diagnosis should be considered in some cases where a mass is closely related to a valve. An

example of valvular pathology mimicking a cardiac mass is caseous mitral annular calcification, where an ovoid mass of caseous calcifications develops in close proximity to the mitral annulus as a result of liquefactive necrosis of mitral valvular calcifications (Fig. 16.4).

280

a

Fig. 16.3 A 44 year-old female with shortness of breath. The sagittal, minimum intensity projection (MinIP) view shows a vermiform, lowattenuation filling defect (black arrow, a), representing a thrombus in the right ventricle, which had likely migrated from the lower extremities

J.E. Shriki et al.

b

or from the pelvic venous system. MinIP views are useful in demonstrating low attenuation structures, when surrounded by relatively high attenuation. A transverse view on the same study shows multiple, separate pulmonary emboli (white arrows, b)

Lesion Morphology Masses should also be described as intramural (within the myocardial wall) or intracameral (within the cardiac chamber). Metastatic and malignant primary tumors usually have a significant intramural component or a very broad-based attachment to the wall of the myocardium, whereas benign masses are more commonly pedunculated and intracameral, often having a narrow attachment. Masses arising from the myocardium tend to have more obtuse angles with the endocardial surface, whereas masses within chambers or with pedunculated attachments tend to have more acute angles with the endocardium. This rubric is commonly helpful in identifying pedunculated masses as benign. Thrombi which are adherent to the internal wall of the ventricle are, however, an important exception to this rule. Lesion shape is less helpful as both benign and malignant masses may be lobulated or appear round.

Fig. 16.4 An 81 year-old male with caseous mitral annular calcifications. A mass was seen near the region of the mitral annulus on echocardiography. CT was performed for further evaluation. A non-contrast, transverse image shows the classic morphology of caseous mitral annular calcifications, with central homogeneous hyperattenuation representing liquified calcifications and denser, peripheral shell-like calcifications (white arrow)

Attenuation Attenuation can be characterized by Hounsfield unit (HU) measurement. Care should be taken to ensure that cardiac gating is adequate, as the presence of motion may alter or artifactually elevate measured attenuation. Attenuation

16

Assessment of Cardiac and Thoracic Masses

281

values from −100 to −10 HU are generally associated with fatty masses such as intracardiac lipomas or lipomatous hyperplasia of the interatrial septum. Frequently, measurement of attenuation is not needed. For example, fatty intracardiac masses can be compared to the attenuation of subcutaneous or mediastinal fat. Cystic masses will tend to have attenuation values between −10 and 10 HU. Calcifications have an attenuation value of 130 HU or greater. Coarse calcifications may be seen in myxomas, although many other lesions may calcify, including some thrombi and many treated metastases. Attenuation relative to muscle or specifically myocardium is frequently used to describe lesions as hypoattenuating or hyperattenuating. Frequently, attenuation relative to the blood pool is also described, although it should be noted that patients with anemia may have depressed pre-contrast attenuation values within vascular structures.

Lesion Number

Enhancement

Thrombi

Enhancement should be reported with respect to the degree of enhancement and to the phase at which enhancement is seen. Lesions which show no or minimal enhancement are more likely to be benign. This is true of thrombi, which usually show no enhancement. Myxomas usually show minimal or mild post-contrast enhancement, particularly in the arterial phase of contrast administration, when CCTA imaging is performed. Angiosarcomas, the most common malignant primary neoplasm of the heart, may have very avid enhancement, to the extent that the borders of these masses may be indistinguishable from contrast within the chamber of the heart. Other neoplastic lesions, including metastases, may show more modest enhancement.

Thrombi are the most common cause of intracardiac masses. On pre-contrast CT, thrombi may either present as hypoattenuating or hyperattenuating masses relative to blood pool (Figs. 16.6 and 16.7). Attenuation relative to blood pool is influenced by the patient’s hematocrit, since more anemic patients will have relatively lower attenuation of blood pool. The degree of attenuation within a thrombus may also be dependent on thrombus age. Most thrombi will show no enhancement after administration of contrast. However, some chronic thrombi, described as being more organized, have been reported to have some peripheral enhancement after contrast administration [29]. This is most commonly seen in the setting of chronic thrombi adherent to the wall, and has been reported mostly on MRI. On CCTA, essentially no contrast enhancement will be shown within thrombi. In the case of small thrombi, Hounsfield units may be elevated after administration of contrast when comparison between pre- and post-contrast CCTA is made, although this is more likely related to pseudoenhancement, whereupon beam hardening effects cause false elevation of attenuation values due to adjacent hyperattenuating structures or contrast. Pseudoenhancement tends to occur in lesions less than 1 cm in size. On MRI, thrombi are most commonly dark on all sequences. Thrombi can also be recognized by the characteristic locations in which they occur, including at the left ventricular apex and in the left atrial appendage. Ventricular thrombi can be recognized by their characteristic location, most commonly at the apex of the left ventricle (Fig. 16.8). Morphologically, they may either present as one or many ovoid structures within the chamber or may appear pedunculated (Fig. 16.9). Many thrombi may also

Involvement of Other Vascular Structures Numerous masses may invade the heart from the great vessels. Tumors of the upper abdomen may grow into the right atrium via the inferior vena cava (Fig. 16.5). Hepatocellular carcinoma, adrenocortical carcinoma, and renal cell carcinoma are among the most common tumors of the upper abdomen to invade into the right atrium. Bronchogenic carcinomas may invade into the heart through the pulmonary veins and present as a left atrial mass. Mediastinal tumors and bronchogenic carcinomas that involve the mediastinum may extend into the heart via the superior vena cava. Tumors of the mediastinum may grow directly into the heart with external myocardial invasion. Thrombi along catheters may track along venous structures, most commonly the superior vena cava.

Multiple lesions are more likely to be due to metastatic disease or to multiple thrombi. Metastatic disease typically appears as multiple lesions in the myocardial wall in different locations. Multiple thrombi may be encountered as well, especially when masses are located in characteristic locations, such as the left atrial appendage or the left ventricular apex.

Commonly Encountered Masses Although a complete discussion of all cardiac masses is beyond the scope of this chapter, familiarity with the most common causes of cardiac masses assists in arriving at the correct diagnosis.

282

J.E. Shriki et al.

a

b

c

Fig. 16.5 A 24 year-old male with retroperitoneal malignant germ cell tumor. Post-contrast CT images are obtained of the abdomen in the portal venous phase and are shown from caudal (a) to cranial (c). There is extensive left periaortic lymphadenopathy, with invasion of tumor into the left renal vein (black arrow on a). There is also extension of tumor

into the inferior vena cava and right atrium (black arrows on b and c, respectively). Note other sites of metastatic disease including a right lower lobe metastasis (white arrow, c) and a left retrocrural lymph node (white arrowhead, b). This tumor exhibits a common appearance of metastatic disease from germ cell tumor with low internal attenuation

be flat and layered against the endocardial surface of the left ventricular wall. Association with an underlying wall motion abnormality, such as an area of aneurysm formation or an area of infarction, is also an important hint to the correct diagnosis. Ventricular thrombi have been reported in up to one third of transmural infarctions [30], and are associated much more commonly with apical and anterior

infarctions, in comparison to inferior infarctions [30, 31]. Visualization of multiple thrombi is not uncommon in the post-infarction setting. Atrial thrombi can be very problematic to confidently diagnose, particularly when present in the left atrial appendage, where contrast opacification is often nonuniform. Patients at risk for atrial thrombi commonly have

16

Assessment of Cardiac and Thoracic Masses

283

a

b

Fig. 16.6 A 56 year-old male with rheumatic heart disease and a hyperattenuating left atrial appendage thrombus. Oblique MPR views are shown from a non-contrast scan (a, b). A focal mass with hyperattenuation is present in the left atrial appendage, which was also seen on

a

echocardiogram (not shown) and was consistent with an atrial thrombus (white arrows). Note the presence of atrial wall calcifications, which are encountered commonly in the setting of rheumatic heart disease. The left atrium is massively enlarged and forms the right heart border (a)

b

Fig. 16.7 A 62 year-old male admitted with a recent myocardial infarction with a hypoattenuating left ventricular thrombus. A transverse, non-contrast view (a) obtained to evaluate the extent of pleural effusion demonstrates an area of low attenuation at the left ventricular apex

(white arrow). A thrombus was suspected. A repeat scan 1 week later obtained with a small amount of intravenous contrast (b) demonstrates clear delineation of the large thrombus at the left ventricular apex (white arrow)

284

a

J.E. Shriki et al.

b

Fig. 16.8 A 45-year-old woman with a recent myocardial infarction. Four-chamber (a) and two-chamber (b) views from a cardiac CT show a left ventricular apical aneurysm with thrombus

a

b

Fig. 16.9 A 55 year-old male with a mass at the ventricular apex on transthoracic echocardiogram. Images from a cardiac CT scan obtained in the two-chamber (a) and short axis (b) planes are shown. A mobile,

intracameral, low attenuating, non-enhancing, apical mass is seen, consistent with a thrombus. A small area of apical septal late gadolinium enhancement was demonstrated on the patient’s CMR (not shown)

enlarged atria with heterogeneous enhancement as a result of circulatory stasis within the left atrium. This smokelike enhancement can be especially prominent in the left atrial appendage, and, in patients with severe atrial enlargement, this poor opacification of the chamber and

appendage may make exclusion of thrombus very difficult (Fig. 16.10). As a result, transesophageal echocardiography is still considered the gold standard for the evaluation of a thrombus in the left atrium or left atrial appendage. Imaging protocols with a delayed phase or with the patient

16

a

Assessment of Cardiac and Thoracic Masses

285

b

Fig. 16.10 A 47 year-old female with atypical chest pain. Orthogonal MPR views of the left atrium from a CCTA (a, b) demonstrate left atrial enlargement with heterogeneous enhancement of the left atrial

appendage. No thrombus was identified on echocardiography. Heterogeneous opacification of the left atrial appendage is a significant pitfall in the identification or exclusion of atrial thrombi on CCTA

ment and dysfunction. Right atrial thrombi may form in the atrial appendage, and can be difficult to delineate due to the inherently heterogeneous opacification of the normal right heart (Fig. 16.12). There are however lower risks of thrombus development in the right atrial appendage, since this structure tends to be flatter and more broadbased, compared to the left atrial appendage, which is a more lobulated or tubular structure and has a neck. These morphologic differences make the left atrial appendage more prone to the development of thrombi compared to the right atrial appendage.

Metastatic Disease

Fig. 16.11 A 48 year-old female with left atrial enlargement and atrial fibrillation. On a contrast-enhanced CT scan, a filling defect is clearly identified in the left atrial appendage (white arrowhead). When such a defect is clearly delineated by contrast as in this case, thrombus can be more definitively identified

in the prone position have been reported as techniques for improved opacification of the atrial appendage, although these techniques are not yet widely employed [32, 33]. When a non-enhancing filling defect is clearly delineated by contrast, however, a left atrial appendage thrombus can be more easily diagnosed (Fig. 16.11). Differentiation of atrial thrombus from other cardiac masses is usually made on the clinical basis in patients with known atrial enlarge-

Metastatic disease is the most common cause of a malignant neoplastic mass in the heart. Metastatic disease has been reported as more common in the right heart, but, as previously mentioned, this may be due to the earlier recognition of metastatic lesions in the right ventricle, since the wall is thinner than that of the left ventricular wall. Most metastatic lesions are isoattenuating to the myocardium on pre-contrast CT imaging. Contrast enhancement of metastatic lesions is somewhat variable depending on the degree of vascularity of the neoplasm. Most metastases enhance less than the myocardium initially after administration of contrast, but will slowly accumulate and retain contrast. Metastases may also show retention of contrast on latephase imaging. This pattern of enhancement may be less well-characterized on CT, since multiphasic imaging with

286

a

J.E. Shriki et al.

b

Fig. 16.12 A 38 year-old female with recent resection of gastric carcinoma. Transverse (a) and four-chamber (b) views are shown from a CT scan of the chest. There is some heterogeneous opacification of the right atrium with some streaking, although the thrombus can be seen through these artifacts (black arrowheads, (a) and (b)). The patient also had a

large parenchymal hepatic hematoma, related to her recent surgery, which may have contributed to the development of a right atrial thrombus. CMR (not shown) demonstrated features consistent with a thrombus, and this structure resolved on subsequent studies

CT is uncommonly performed (Fig. 16.13). It should also be noted that the optimal phase for CCTA during coronary arterial enhancement is earlier than the optimal phase for demonstrating myocardial wall enhancement. As a result, the differential enhancement between a metastatic lesion and normal myocardium may not be as clearly seen on CCTA (Fig. 16.14). The most common morphology of cardiac metastatic disease is an intramural mass or a mass with a broad-based attachment, in contradistinction to benign masses, which tend to be intracameral and have a narrow attachment. Many metastases which invade the heart from the adjacent mediastinal or pericardial spaces may, however, involve the epicardium first, and subsequently invade into the myocardium. Metastatic disease to the heart is found in up to 10 % of patients with a primary malignancy at the time of autopsy [34]. Although numerous neoplasms have been reported to be metastatic to the heart, the lung is the most common site of a primary tumor, occurring in up to 36.7 % of patients [35]. Melanoma is however, an important source of hematogenous spread of disease to the heart from a distant primary site [36]. An example of metastases involving the right ventricle is shown in Fig. 16.15.

Myxomas characteristically occur in the atria, and are more commonly left atrial rather than right atrial, with a reported predominance of 80 % in the left atrium compared to 20 % in the right atrium. Masses which arise in the atria may also prolapse into the ventricular chambers [14]. Myxomas have also, however, been reported to occur in both ventricles. The most common imaging appearance is that of a lobulated mass with pre-contrast hypoattenuation relative to blood pool and relative to myocardium. Masses are commonly lobulated in appearance and predominantly intracameral. The most common site of attachment for either right or left atrial myxomas is at the fossa ovalis, a feature which can be helpful in arriving at the correct diagnosis [9]. Correct identification of the site of attachment is also helpful in presurgical planning. Atrial myxomas commonly demonstrate punctate or coarse calcifications, which is also useful in establishing the diagnosis. Pre-contrast hypoattenuation is commonly seen relative to blood pool and normal myocardium (Fig. 16.16). Rarely, atrial myxomas may be diffusely and densely calcified [17]. The classic triad of clinical symptoms reported with myxomas includes constitutional symptoms, manifestations of obstructive valvular disease, and embolic phenomenon. Constitutional symptoms include fever, malaise, weight loss, and anemia, among others. These symptoms are likely related to an autoimmune response initiated by the tumor [37]. Cardiac-related symptoms of atrial myxomas vary depending on the chamber of involvement. Atrial myxomas have been commonly reported to mimic mitral valve disease

Myxomas Myxomas are the most common benign neoplasm of the heart and comprise 50 % of all primary cardiac masses.

16

Assessment of Cardiac and Thoracic Masses

a

287

b

c

Fig. 16.13 A 45 year-old female with metastatic melanoma. A transverse view from a contrast-enhanced CT scan (a) shows irregular thickening of the ventricular walls and a moderate-sized pericardial effusion. The ventricular metastases are not well-delineated due to the early phase of contrast administration, which was in part, due to the patient’s poor cardiac function. A post-contrast CMR sequence obtained at 70 s

after intravenous injection of gadolinium (b) demonstrates areas of relatively decreased enhancement related to the perfused myocardium (white arrows). A delayed, 4-chamber view obtained 10 min after contrast administration (c) demonstrates late enhancement within the cardiac metastases (white arrows)

and rheumatic heart disease by clinical presentation [38]. Involvement of other valves may, however, produce manifestations of aortic, pulmonic, or tricuspid valvular disease. Embolization is another common feature of myxomas, and may occur to either the pulmonic or systemic circulation, depending on the chamber of involvement. Up to 35 % of left atrial myxomas and up to 10 % of right atrial myxomas may

embolize, although this difference in embolization rate could be related to the more apparent manifestations of systemic emboli [39]. A genetic predisposition to myxomas has been postulated and suggested by case reports of families with multiple members with myxomas and in patients with several myxomas [40]. Notably, Carney’s Syndrome may be associated

288

Fig. 16.14 A 56 year-old male with high-grade urothelial malignancy and cardiac metastases. Transverse views from an arterial phase of a post-contrast CT scan (top row) faintly show metastases to the left ventricular myocardium (white arrows). These areas of hypoenhancement

a

Fig. 16.15 A 48 year-old female with right ventricular metastases due to thymic carcinoma. Transverse (a) and coronal (b) views from a postcontrast CT scan show large masses arising from the wall of the right

J.E. Shriki et al.

are better seen in the portal venous phase images (bottom row), where the myocardium is more well-enhanced. Note that other metastases are also better seen on the later phase study, including pleural metastases (black arrows)

b

ventricle, with broad-based attachments to the ventricular wall, consistent with metastatic disease

16

Assessment of Cardiac and Thoracic Masses

a

289

b

Fig. 16.16 A 52 year-old male with treated colorectal carcinoma and a left atrial mass. A pre-contrast CT scan (a) demonstrates the low attenuation left atrial mass. The post-contrast study (b) shows no

enhancement, and delineates the pedunculated nature of the mass, which arises from the region of the fossa ovalis

with atrial myxomas, occurring in two thirds of patients, in addition to other manifestations including mammary myxoid fibroadenomas, pigmented cutaneous lesions, endocrine disorders, testicular tumors, and schwannomas [41].

tumors may be difficult to delineate from the ventricular chamber on later phases due to bright enhancement. Venous lakes and linear vascular structures within masses may be seen, resulting in what has been likened to a “sunray pattern” [43]. Two morphological appearances of angiosarcomas have been reported, including a focal mass arising from the myocardium itself or a diffuse infiltrating process involving the myocardium and pericardium [44, 45]. Undifferentiated sarcomas are tumors with no specific histological staining patterns. The nature and definition of tumors in this category has changed over time as histological techniques have improved. Similar to angiosarcomas, these tumors may either present as a focal mass or as a diffusely infiltrative myocardial and pericardial process. The common site of origin is the left atrium, with a predisposition reported at 80 % [13]. A propensity for valvular involvement has also been reported [46–48]. Rhabdomyosarcomas are very uncommon in adults, but are the most common form of cardiac sarcomas and the most common primary cardiac malignancy in pediatric populations [13]. Embryonal rhabdomyosarcomas occur in pediatric patients, whereas tumors in adults tend to be more pleomorphic [49]. Osteosarcomas of the heart are rare neoplasms, and are distinguished by their propensity to form

Cardiac Sarcomas Cardiac sarcomas comprise the most common primary malignant tumors of the heart, but are a rare entity overall, with a prevalence at autopsy as low as 0.0001 % [42]. Metastases to the heart outnumber malignant primary lesions by a ratio of 20–40 to 1. Among subtypes of sarcoma, angiosarcomas are most common, comprising approximately 37 %. This tumor subtype in particular tends to occur commonly in the right atrium. Other subtypes tend to arise most commonly from the left atrium, although all types of sarcomas may occur in any chamber [13]. For most cardiac sarcomas, survival is reported as very poor, with metastases commonly detected shortly after clinical presentation [10]. On CCTA and CMR, angiosarcomas may show areas of hemorrhage and necrosis and may appear heterogeneous. Avid enhancement is commonly seen (Fig. 16.17), and

290

J.E. Shriki et al.

a

b

Fig. 16.17 (a, b) A 33 year-old female with angiosarcoma. Sequential pre-contrast, early, and late contrasted images demonstrate vascular enhancement within the angiosarcoma. Note the enhancing right lower lobe pulmonary nodule (arrowhead), consistent with a metastasis

dense calcifications [50], although some tumors of this type may demonstrate only minimal calcification [13]. Other tumor subtypes include leiomyosarcoma, fibrosarcoma, and liposarcoma, although these are even rarer than the aforementioned neoplasms.

Cardiac Lymphoma Cardiac lymphoma is a very rare entity, and in a series of 533 cardiac tumors and cysts, it accounted for only 1.3 % of tumors [11]. These tumors are rare since there are no true

intracardiac lymph nodes. Tumors likely arise from primitive, totipotential mesenchymal cells, and usually consist of high grade B-cell lymphomas. Strictly defined, cardiac lymphoma includes lymphoma involving the heart and pericardium without other areas of lymphomatous involvement. Anecdotal reports suggest that there is increased risk for cardiac lymphoma in AIDS and in other immune deficiency states [51]. Given the rarity of this entity, the radiologic findings are not well-established, although reports indicate that tumors are usually relatively isoattenuating on CT and isointense on CMR, with heterogeneous enhancement after contrast administration [52].

16

Assessment of Cardiac and Thoracic Masses

Lipoma Cardiac lipomas are the second most common cause of a benign cardiac neoplasm, after myxomas. Lipomas are easily recognized as benign by the homogeneously low, precontrast attenuation consistent with fat, which demonstrate essentially no enhancement after contrast administration. Tumors are soft and may be large at the time of initial diagnosis. Symptoms are usually due to mass effect, although commonly cardiac lipomas are detected incidentally and prior to onset of clinical manifestations. Some tumors may encase coronary arteries, resulting in mass effect and displacement, making resection difficult [53]. Although there is seldomly diagnostic uncertainty, some entities may mimic lipomas, including lipomatous hypertrophy of the interatrial septum (Fig. 16.18). Rarely, lipomatous metaplasia within chronic myocardial infarctions may be misdiagnosed as a lipoma [54].

Papillary Fibroelastoma Papillary fibroelastomas, also known as Lambl’s excrescences, are avascular masses comprised of fronds of dense connective tissue. These masses may be either reactive in nature or may be related to a hamartoma [55]. The true prevalence of this entity is not known, and these tumors have been postulated to be under-recognized and under-diagnosed due to their small size. Most sources refer to these lesions as

a

Fig. 16.18 A 78 year-old male with lipomatous hypertrophy of the interatrial septum. Images from a CT scan of the chest obtained in the transverse plane (a) and in the short axis plane of the heart (b) show

291

the third most common benign neoplasms behind myxomas and lipomas. Ninety percent of papillary fibroelastomas occur on valves, with the aortic valve being the most common location. When associated with the atrioventricular valves, these tumors tend to occur along the atrial side, which may help to differentiate these lesions from thrombi [56]. Associated valvular dysfunction is common, although many of these lesions are detected incidentally [57]. An additional, common presentation is the occurrence of embolic phenomenon to the systemic or pulmonary circulation [58]. These tumors are, however, uncommonly reported on CCTA, and as a result no characteristic CCTA features of this entity have emerged.

Pediatric Cardiac Masses In pediatric patients who present with cardiac masses, a separate set of diagnostic possibilities should be considered. Pediatric patients will less commonly present for CCTA evaluation, due to radiation concerns. Additionally, since there is sparse literature on CCTA in pediatric patients, the typical appearances of many masses are difficult to delineate. However, some familiarity with masses which may present in pediatric patients is useful to physicians involved in cardiac imaging. In infants and children, the most common masses encountered are rhabdomyomas. These masses tend to occur in the walls of the ventricles, and the vast majority of these masses

b

lipomatous hypertrophy in the wall of the interatrial septum (white arrows). Note the characteristic sparing of the region of the fossa ovalis (black arrow)

J.E. Shriki et al.

292

are multiple. A strong association with tuberous sclerosis is present, although many patients do not manifest other signs of tuberous sclerosis until many years later. Spontaneous regression of tumors is common, although the initial clinical presentation may be severe. When these lesions regress, they may leave behind foci of myocardial fat attenuation [59]. Cardiac fibromas constitutes the second most common cause of pediatric cardiac masses, and usually are seen as large, solitary lesions. Some of these may grow to an enormous size. Symptoms may include heart failure, chest pain, arrhythmias, and sudden cardiac death [60]. Myxomas are very rare in the pediatric population, but have been reported in large series in older pediatric patients and adolescents [17]. As described earlier, the most common malignant neoplasm in pediatric patients is a cardiac rhabdomyosarcoma, which often has embryonal features at histology. Additional rare tumors are encountered in pediatric populations including cardiac angiomas, cardiac teratomas, and Purkinje cell tumors.

5.

6.

7.

8. 9. 10. 11.

12. 13.

Mimics of Cardiac Masses

14.

Several normal structures in the heart may mimic a mass even to an experienced reader. Misidentification of normal structures as masses can lead to unnecessary biopsies, surgeries, and subsequent morbidity. Notably, in the right atrium, a prominent crista terminalis may mimic a cardiac mass. When a small and reticulated mass is present along this structure in the right atrium, a network of Chiari may be present. Uncommonly, prominence of the Eustachian valve or juxtacaval lipomatous tissue may mimic a lower right atrial mass. Usually, the contrast timing for the right ventricle is such that some saline is being injected at the time of scanning of the heart. However, in some cases, unopacified blood flow entering the right atrium from the inferior vena cava may simulate a mass when the right atrium is otherwise filled with contrast. In the left atrium, the “coumadin ridge,” located between the atrial appendage ostium and the superior pulmonary vein, may appear mass-like.

15.

16.

17.

18.

19.

20. 21.

22.

References 1. Desjardins B, Kazerooni EA. ECG-gated cardiac CT. Am J Roentgenol. 2004;182:993–1010. 2. Newhouse JH, Murphy RX. Tissue distribution of soluble contrast: effect of dose variation and changes with time. Am J Roentgenol. 1981;136:463–7. 3. Cademartiri F, Mollet N, van der Lugt A, et al. Non-invasive 16-row multislice CT coronary angiography: usefulness of saline chaser. Eur Radiol. 2004;14:178–83. 4. Awai K, Hiraishi K, Hori S. Effect of contrast material injection duration and rate on aortic peak time and peak enhancement

23.

24.

25. 26.

at dynamic CT involving injection protocol with dose tailored to patient weight. Radiology. 2004;230(1):142–50. Cademartiri F, Nieman K, van der Lugt A, et al. Intravenous contrast material administration at 16-detector row helical CT coronary angiography: test bolus versus bolus-tracking technique. Radiology. 2004;233:817–23. Bae KT, Tran HQ, Heiken JP. Uniform vascular contrast enhancement and reduced contrast medium volume achieved by using exponentially decelerated contrast material injection method. Radiology. 2004;231:732–6. Boyd DB. Computerized transmission tomography of the heart using scanning electron beams. In: Higgins CH, editor. Computed tomography of the heart and great vessels. Mount Kisco/New York: Futura Publishing Company; 1983. p. 45–55. Newhouse JH. Fluid compartment distribution of intravenous iothalamate in the dog. Invest Radiol. 1977;12:364–7. Prichard RW. Tumors of the heart. Arch Pathol. 1951;51:98–128. Glancy DL, Morales JB, Roberts WC. Angiosarcoma of the heart. Am J Cardiol. 1968;21:413–9. McAllister HA, Fenoglio Jr JJ. Tumors of the cardiovascular system: atlas of tumor pathology, vol. 2. Washington, DC: Armed Forces Institute of Pathology; 1978. Lund JT, Ehman RL, Julsrud PR, et al. Cardiac masses: assessment by MR imaging. AJR Am J Roentgenol. 1989;152(3):469–73. Burke AP, Cowan D, Virmani R. Primary sarcomas of the heart. Cancer. 1992;69:387–95. Tazelaar HD, Locke TJ, McGregor CG. Pathology of surgically excised primary cardiac tumors. Mayo Clin Proc. 1992;67(10):957–65. Lam KY, Dickens P, Chan AC. Tumors of the heart. A 20-year experience with a review of 12,485 consecutive autopsies. Arch Pathol Lab Med. 1993;117(10):1027–31. Marx GR. Cardiac tumors. In: Emmanouilides GC, Gutgesell HP, Riemenschneider TA, Allen HD, editors. Moss and Adams heart disease in infants, children, and adolescents: including the fetus and young adult, vol. 2. 5th ed. Baltimore: Williams & Wilkins; 1995. p. 1773–86. Burke A, Virmani R. Tumors of the heart and great vessels: atlas of tumor pathology. Fasc 16, ser 3. Washington, DC: Armed Forces Institute of Pathology; 1996. Takach TJ, Reul GJ, Ott DA, Cooley DA. Primary cardiac tumors in infants and children: immediate and long-term operative results. Ann Thorac Surg. 1996;62(2):559–64. Ludomirsky A. Cardiac tumors. In: Bricker JT, Fisher DJ, editors. The science and practice of pediatric cardiology, vol. 2. 9th ed. Baltimore: Williams & Wilkins; 1998. p. 1885–93. Araoz PA, Eklund HE, Welch TJ, Breen JF. CT and MR imaging of primary cardiac malignancies. Radiographics. 1999;19(6):1421–34. Chiles C, Woodard PK, Gutierrez FR, Link KM. Metastatic involvement of the heart and pericardium: CT and MR imaging. Radiographics. 2000;20:1073–103. Grebenc ML, Rosado de Christenson ML, Burke AP, Green CE, Galvin JR. Primary cardiac and pericardial neoplasms: radiologic– pathologic correlation. Radiographics. 2000;20:1073–103; quiz 1110–1071, 1112. Grebenc ML, Rosado-de-Christenson ML, Green CE, Burke AP, Galvin JR. Cardiac myxoma: imaging features in 83 patients. Radiographics. 2002;2(3):673–89. Piazza N, Chughtai T, Toledano K, et al. Primary cardiac tumours: eighteen years of surgical experience on 21 patients. Can J Cardiol. 2004;20(14):1443–8. Tatli S, Lipton MJ. CT for intracardiac thrombi and tumors. Int J Cardiovasc Imaging. 2005;21(1):115–31. Butany J, Nair V, Naseemuddin A, et al. Cardiac tumours: diagnosis and management. Lancet Oncol. 2005;6(4):219–28.

16

Assessment of Cardiac and Thoracic Masses

27. Leipsic JA, Heyneman LE, Kim RJ. Cardiac masses and myocardial diseases. In: McAdams HP, Reddy GP, editors. Cardiopulmonary imaging syllabus—2005. Leesburg: American Roentgen Ray Society; 2005. p. 1–13. 28. Sparrow PJ, Kurian JB, Jones TR, Sivananthan MU. MR imaging of cardiac tumors. Radiographics. 2005;25:1255–76. 29. Barkhausen J, Hunold P, Eggebrecht HH, et al. Detection and characterization of intracardiac thrombi on MR imaging. AJR Am J Roentgenol. 2002;179:1539–42. 30. Weinreich DJ, Burke JF, Ferrel JP. Left ventricular mural thrombi complicating acute myocardial infarction long-term follow-up with serial echocardiography. Ann Intern Med. 1984;100(6):789–94. 31. Asinger RW, Mikell FL, Elsperger J, Hodges M. Incidence of left ventricular thrombosis after acute transmural myocardial infarction: serial evaluation by two-dimensional echocardiography. N Engl J Med. 1981;305(6):297–302. 32. Hur J, Kim YJ, Nam JE, Choi EY, Shim CY, Choi BW, et al. Thrombus in the left atrial appendage in stroke patients: detection with cardiac CT angiography--a preliminary report. Radiology. 2008;249(1):81–7. 33. Hur J, Kim YJ, Lee HJ, et al. Left atrial appendage thrombi in stroke patients: detection with two-phase cardiac CT angiography versus transesophageal echocardiography. Radiology. 2009;251(3):683–90. 34. Abraham KP, Reddy V, Gattuso P. Neoplasms metastatic to the heart: review of 3314 consecutive autopsies. Am J Cardiovasc Pathol. 1990;3:195–8. 35. Klatt EC, Heitz DR. Cardiac metastases. Cancer. 1990;65:1456–9. 36. MacGee W. Metastatic and invasive tumours involving the heart in a geriatric population: a necropsy study. Virchows Arch A Pathol Anat Histopathol. 1991;419:183–9. 37. Bjessmo S, Ivert T. Cardiac myxoma: 40 years’ experience in 63 patients. Ann Thorac Surg. 1997;63:697–700. 38. Markel ML, Waller BF, Armstrong WF. Cardiac myxoma: a review. Medicine. 1987;66:114–25. 39. Castells E, Ferran V, Octavio-de-Toledo MC. Cardiac myxomas: surgical treatment, long-term results and recurrence. J Cardiovasc Surg. 1993;34:49–53. 40. Carney JA. Differences between nonfamilial and familial cardiac myxoma. Am J Surg Pathol. 1985;9:53–5. 41. Carney JA, Gordon H, Carpenter PC, Shenoy BV, Go VW. The complex of myxomas, spotty pigmentation and endocrine overactivity. Medicine. 1985;64:270–83. 42. McCallister Jr HA. Primary tumors of the heart and pericardium. Curr Probl Cardiol. 1979;4:1–51. 43. Yahata S, Endo T, Honma H, et al. Sunray appearance on enhanced magnetic resonance image of cardiac angiosarcoma with pericardial obliteration. Am Heart J. 1994;127:468–71.

293 44. Bruna J, Lockwood M. Primary heart angiosarcoma detected by computed tomography and magnetic resonance imaging. Eur Radiol. 1998;8:66–8. 45. Jannigan DT, Husain A, Robinson NA. Cardiac angiosarcomas: a review and a case report. Cancer. 1986;57:852–9. 46. Herhusky MJ, Gregg SB, Virmani R, Chun PKC, Bender H, Gray Jr GF. Cardiac sarcoma presenting as metastatic disease. Arch Pathol Lab Med. 1985;109:943–5. 47. Ludomirsky A, Vargo TA, Murphy DJ, Gresik MV, Ott DA, Mullins CE. Intracardiac undifferentiated sarcoma in infancy. J Am Coll Cardiol. 1985;6:1362–4. Abstract 37. 48. Itoh K, Matsumura T, Egawa Y, et al. Primary mitral valve sarcoma in infancy. Pediatr Cardiol. 1998;19:174–7. 49. Hwa J, Ward C, Nunn G, et al. Primary interventricular cardiac tumors in children: contemporary diagnostic and management options. Pediatr Cardiol. 1994;15:233–7. 50. Chaloupka JC, Fishman EK, Siegelman SS. Use of CT in the evaluation of primary cardiac tumors. Cardiovasc Intervent Radiol. 1986;9:132–5. 51. Holladay AO, Siegel RJ, Schwartz DA. Cardiac malignant lymphoma in acquired immune deficiency syndrome. Cancer. 1997;70(8):2203–7. 52. Dorsay TA, Ho VB, Roviera MJ, Armstrong MA, Brissette MD. Primary cardiac lymphoma: CT and MR findings. J Comput Assist Tomogr. 1993;17:978–81. 53. Hananouchi GI, Goff WB. Cardiac lipoma: six-year follow-up with MRI characteristics, and a review of the literature. Magn Reson Imaging. 1990;8(6):825–8. 54. Banks KP, Lisanti CJ. Incidental finding of a lipomatous lesion involving the myocardium of the left ventricular wall. AJR Am J Roentgenol. 2004;182:261–2. 55. Rubin MA, Snell JA, Tazelaar HD, Lack EE, et al. Cardiac papillary fibroelastoma: an immunohistochemical investigation and unusual clinical manifestations. Mod Pathol. 1995;8:402–7. 56. Klarich KW, Enriquez-Sarano M, Gura GM, et al. Papillary fibroelastoma: echocardiographic characteristics for diagnosis and pathologic correction. J Am Coll Cardiol. 1997;30:784–90. 57. Edward FH, Hale D, Cohen A, et al. Primary cardiac valve tumors. Ann Thorac Surg. 1991;52:1127–31. 58. McFadden PM, Lacy JR. Intracardiac papillary fibroelastoma: an occult cause of embolic neurologic deficit. Ann Thorac Surg. 1987;43:667–9. 59. Bosi G, Lintermans JP, Pellegrino PA, et al. The natura history of cardiac rhabdomyoma with and without tuberous sclerosis. Acta Paediatr. 1996;85:928–31. 60. Turi GK, Albala A, Fenoglio Jr JJ. Cardiac fibromatosis: an ultrastructural study. Hum Pathol. 1980;11:577–9.

Part IV CT Vascular Angiography

CT Angiography of the Peripheral Arteries

17

Jabi E. Shriki, Leonardo C. Clavijo, and Gale L. Tang

Abstract

The application of CT angiography to the systemic vascular tree poses a number of unique challenges, but offers the ability to noninvasively depict a wide array of arterial pathology. CT of the peripheral arterial tree also has a number of specific advantages relative to other modalities, including conventional angiography, MRI, and ultrasound. Peripheral CT angiography has particularly important applications in imaging extremities in the setting of acute ischemia or trauma. While some of the skills in performing CT angiography in other body parts are applicable to peripheral CT angiography, several technical considerations should be recognized in incorporating peripheral imaging into a CT angiography practice. Keywords

Peripheral computed tomography • Peripheral vascular ct angiography • Peripheral angiogram • Peripheral vascular disease • Systemic arterial disease • Vascular cta • Peripheral ct angiogram

Introduction CT angiography is a useful modality in imaging the peripheral arterial tree and has become an integral component in many cardiovascular imaging practices. Large portions of the arterial system can be easily imaged with excellent spatial resolution, low radiation dose, and minimal risk to the patient. Peripheral CT angiography has particular advantages as a non-invasive means of depicting the systemic arterial tree and is able to demonstrate a number of

J.E. Shriki, MD (*) Department of Radiology, Puget VA Health System, University of Washington, 1660 S. Columbian Way, Seattle, WA 98101, USA e-mail: [email protected] L.C. Clavijo, MD, PhD, FACC, FSCAI, FSVM Department of Medicine, Division of Cardiovascular Medicine, Department of Clinical Medicine, University of Southern California, Los Angeles, CA, USA G.L. Tang, MD Department of Surgery, University of Washington, Seattle, WA, USA

disease entities. Additionally, the skills of 3-D data manipulation useful in evaluation of coronary arteries and vascular structures elsewhere in the body can be translated into skills in interpreting peripheral CT angiographic studies. Advancements in CT angiography, including the dissemination of multislice, dual source, and dual energy scanners, have made submillimeter isotropic voxel resolution possible, and have enabled more detailed visualization of arterial structures. Simultaneous increases in computational speed and widespread availability of dedicated 3-D software make visualization and evaluation of peripheral arterial anatomy much more facile.

Acquisition and Scanning Techniques Special Considerations Regarding Peripheral CT Angiography Several technical considerations should be recognized in making the transition from cardiac to peripheral vascular CT

© Springer International Publishing 2016 M.J. Budoff, J.S. Shinbane (eds.), Cardiac CT Imaging: Diagnosis of Cardiovascular Disease, DOI 10.1007/978-3-319-28219-0_17

297

298

angiography. First, while the coronary arteries are usually opacified along with the aorta even in the presence of stenoses, the peripheral vasculature may have a variable relationship with aortic opacification. In patients with severe atherosclerotic disease, the presence of stenoses, occlusions, and aneurysms may delay optimal opacification of the peripheral arteries. In thoracic, abdominal, and neuroimaging applications, higher detector row CT scanners provide a number of advantages. However, for peripheral CT angiography, the high speed of scanning with multi-detector row CT scanners may result in outpacing of the bolus of contrast, with images obtained prior to arrival of the contrast bolus into the area of interest. As a result, the timing of peripheral CT angiography has separate considerations that differ from scanning other, more proximal, body parts (Fig. 17.1). Also, the reconstructed field of view for peripheral applications is frequently larger in transverse axial dimension than that employed for cardiac CT. This is because of the wider distribution of peripheral arterial structures in the transverse plane. As a result, images may have lower inplane spatial resolution. This loss of in-plane resolution may be offset by the use of separate, reconstructed fields of view for each lower extremity or for different parts of the anatomy scanned. When imaging the lower extremities, the feet should be kept straight and positioned as closely together as possible, so that the reconstructed field of view closely matches the anatomy being imaged. This can be achieved by securing the feet into a table extension or harness, which is usually attachable or built into the table (Fig. 17.2). In comparison to cardiac CT, there is a longer craniocaudal extent of anatomy imaged with peripheral CT angiography. As a result, data sets may be much larger for comparable slice thickness. For example, whereas a single phase of a cardiac CT reconstruction at 0.5 mm may comprise 1–200 or more images, a data set from a peripheral CT angiogram of the lower extremities might include several thousand images, due to the craniocaudal extent of imaging from the diaphragm to the toes. As a result, some readers prefer thicker slices or coronal plane images for initial evaluation, and reserve the use of thin slices for a more limited, adjunctive role in problem solving. Alternatively, most 3-D workstations have options for subvolume selection, which enables evaluation of the arterial tree in an incremental fashion, allowing larger data sets to be evaluated, with enhanced multiplanar reformatting capabilities.

Dual Energy and Dual Source CT The advent of dual source and dual energy scanning has enabled a new set of advantages of CT imaging for the peripheral vascular tree. It should be noted that the descriptors of “dual energy” and “dual source” CT are sometimes used

J.E. Shriki et al.

Fig. 17.1 A 3-D reconstruction demonstrating that the run-off vessels are less well opacified than the popliteal arteries, likely related to slight outpacing of the contrast bolus by the speed of the scanner

interchangeably, although there are differences between the terms. Dual source CT is a technique of scanning with orthogonally positioned CT acquisition systems (including x-ray source and x-ray detector) mounted to the same gantry. This enables fast and essentially simultaneous acquisition of scan data with two separate energies. The main advantage of dual source CT is that scanning at two different energies can

17 CT Angiography of the Peripheral Arteries

Fig. 17.2 Reconstructed volume projections can demonstrate the soft tissue anatomy. The lower extremities should be closely apposed to one another in order to minimize the dimensions of the reconstructed field of view. In this case, this is achieved by use of a table extension (white arrow)

be obtained rapidly, resulting in excellent temporal resolution with minimal mismatch between acquisitions. However, dual source CT scanning is one means of obtaining scan data at two different energies, and is therefore, best considered a subtype of dual energy CT. Other means of scanning at two energies are available, including rapid switching of kilovolt potentials (kVp) of the x-ray generating tube, or selective detector arrays which are sensitive to different types of radiation. Dual energy scanning using sequential scanning with two different energies has been a research tool since soon after the advent of CT [1]. The development of more advanced CT acquisition techniques has enabled dual energy CT using dual source CT scanners and other techniques to be commercially available since 2006 [2]. A further, more extensive discussion of the physics of dual energy and dual source CT is beyond the scope of this chapter, but typically, the two energies during which scanning is performed are 80 and 140 kV. There are several potential advantages of dual energy CT, including plaque characterization and improvement of contrast visualization. In peripheral CT angiography, the main advantage of scanning at two energies is that higher energy scan data can be obtained, resulting in selective subtraction of higher attenuation materials, such as calcium or stent material [3]. As a result, the contrast column within the vessel can be depicted more easily, without obscuration or blooming from high attenuation calcium or stent wall (Fig. 17.3).

Contrast Administration As with imaging other vasculature, rapid rates of contrast administration are critical in obtaining optimal peripheral vascular opacification. Consequently, a more central, large bore, venous access line (usually consisting of an 18-gauge catheter in the antecubital fossa) is highly preferable to a

299

smaller or more peripheral venous access site. Most studies for peripheral CT angiography report contrast rates of 3.5– 4.0 ccs per second as optimal [4–6]. Since the area scanned in imaging the peripheral arterial tree is significantly larger in craniocaudal extent compared to the coronary tree, a more prolonged bolus of contrast, with a slightly slower rate of delivery is preferable to ensure homogeneous, persistent, bright opacification of the peripheral arteries. This injection rate is slightly slower than the rate employed for imaging the heart, which may be as high as 5–6 ccs per second [7], since the aim of peripheral CT angiography is a sustained peak of bright opacification, whereas in coronary CT, prolongation of peak contrast opacification is a less important factor. Patients are usually given a formulation of intravenous contrast with 300–400 mg of iodine per mL, with 120–180 ccs of contrast given depending on each patient’s body surface area [8]. The total amount of contrast, however, may be reduced when scanners with higher numbers of detectors are used [9, 10]. Higher iodine concentrations have been demonstrated to have higher attenuation levels when the aortic enhancement is compared [11]. Larger amounts of contrast are needed in patients who are taller or are more obese [8]. Administration of a saline chaser is useful in ensuring a higher degree of opacification, and also in prolonging the plateau of attenuation once the peak is reached. A saline chaser is also useful in diminishing the total amount of contrast needed for optimal opacification [12, 13]. Saline injection can also clear residual contrast from the central venous system. Central venous? stasis of contrast can impede imaging of the central upper extremity arteries due to streaking as a result of dense contrast in the superior vena cava, brachiocephalic veins, or other venous structures. When imaging the upper extremities, contrast injection should be made via the extremity contralateral to the area of interest to avoid this pitfall. Optimal timing for acquisition varies significantly in each patient. In patients with normal cardiac function, a rapid acquisition of images may outpace the bolus of contrast. In patients with depressed cardiac function or arterial pathology, the scanning time should be prolonged to ensure scanning is not performed before arrival of the contrast bolus [14]. At our institution, in patients with suspected atherosclerotic disease, the lower extremities are scanned twice, with a second acquisition beginning just above the knees and timed immediately after the first acquisition. This second acquisition enhances visualization of arterial structures, although there may be significant venous contamination at the time of a second scan, which may make 3-D reconstructions somewhat difficult to evaluate. Several techniques for ensuring optimal arterial timing may be employed. A timing bolus utilizes injection of a small amount of contrast with serial images through a region of interest (ROI) in order to predict the timing of bolus arrival. This is somewhat problematic in evaluating the

300

a

J.E. Shriki et al.

b

Fig. 17.3 Frontal, thick volume, maximum intensity projection CT images are shown with bone removal. The scan is obtained with dual energy CT, enabling subtraction imaging. Images are shown before (a) and after (b) subtraction of high attenuation materials, including the

bilateral common iliac artery stents (white arrows) and calcified plaque in the aorta (white arrowheads). Subtraction of high attenuation structures, including stents and vascular calcifications is one of the advantages of dual energy, dual source CT

peripheral arterial tree, since different portions may be opacified at different times, depending on the degree of upstream disease. This technique also necessitates two separate injections with an initial, small bolus. Because only a small, initial dose of contrast is used though, the total amount of contrast is generally not significantly impacted. This technique may also help in preparing the patient for the clinical, physiologic manifestations such as sensory warmth and a metallic taste which commonly ensue after contrast administration. Alternatively, bolus tracking can be performed with the main contrast injection. With this technique, an ROI within the aorta is serially scanned during the injection of contrast. When

the attenuation value reaches a particular, preset threshold, scanning of the remainder of the field of view is initiated. At our institution, for peripheral CT angiography of the lower extremities, an attenuation value of 180 Hounsfield units (HU) is employed, and the ROI is placed in the infrarenal aorta. Alternatively, the ROI can also be placed in the lower extremity arteries, such as the femoral arteries. This technique has the pitfall of being affected by patient motion, and may require some technologist expertise in identifying the vessel. When bolus triggering is used, the ROI is generally positioned to include approximately half of the diameter of the vessel. Different vendor-specific protocols are available for automated

17 CT Angiography of the Peripheral Arteries

contrast monitoring. For slower scanners, a lower threshold (100 HU) may be used to ensure that the speed of scanning matches arrival of the contrast bolus [15]. A significant limitation of the bolus tracking technique is that the ROI may be placed within an area of thrombus or in the false lumen of a dissection. If this occurs, opacification within the ROI may not be achieved, and scanning might be incorrectly delayed. Preset timing of scanning uses a fixed time interval between initiation of contrast administration and scanning. This is less commonly employed at most institutions, especially for imaging peripheral arteries. This technique may be especially problematic in patients with atherosclerotic disease and in patients with low cardiac output, where the bolus will be circulated through the arteries more slowly. For the upper extremities, scanning may be initiated 20 s after the start of contrast injection. For the lower extremities, scanning may be initiated 50 s after the beginning of injection [16].

301

Several studies have demonstrated an excellent accuracy of CT in comparison to conventional digital subtraction angiography (DSA), based solely on evaluation of transverse

images [17, 18]. In many early studies, only the transverse axial data set was evaluated. The transverse plane of the body is in a relatively perpendicular axis to the long arteries of the extremities, resulting in views which approximate the short axis plane of much of the vascular tree with minimal technical manipulation of data sets. For a more accurate interpretation, final review of studies at a dedicated 3-D workstation is commonly employed. Key images for demonstrating stenoses or other vascular pathology are subsequently also sent to archiving and communication systems (PACS) to illustrate important findings. In addition to the evaluation of transverse axial data sets, review of long and short axis planes utilizing multiplanar reformatted views (MPR), thick maximum intensity projection views (MIP) (Fig. 17.4), and curvilinear plane reformatted views (CPR) (Fig. 17.5) result in a more thorough assessment of the peripheral vascular tree and in improved sensitivity and specificity for depicting disease [19]. Review of the transverse axial views is the usual starting point for most readers. Reformatting data along the plane of the vessel utilizing MPR views introduces few artifacts, as long as scans are obtained using isotropic voxels. MIP views generally demonstrate the higher attenuation values within a thick slab of the data set, and are useful for demonstrating the

Fig. 17.4 A reformatted view through the radial artery is shown (a). On the progressively thicker maximum intensity projection (MIP) views obtained with a thickness of 2 cm (b) and 4 cm (c), a greater

length of the arterial anatomy is demonstrated. MIP views are also useful for demonstrating high attenuation structures such as bones and stents

Techniques for Interpreting Studies

302

J.E. Shriki et al.

Fig. 17.5 Curved planar reformatted views show the long axis, orthogonal (white arrows) and short axis (closed arrowhead) of arterial structures. This is contrasted with the MIP MPR view (open arrowhead)

course of a tortuous vessel. MIP views are also useful for demonstrating other high attenuation structures such as stents (Fig. 17.6a), surgical clips, calcifications, and osseous structures. CPR images introduce some potential artifacts, as computer algorithms select the center line to be followed. Frequently, computer-generated center lines may drift into an area of calcification in the wall, and may make a stenosis appear more severe. User-directed CPR images may be created, but are commonly time consuming and require some expertise to generate. Unlike evaluation of the coronary arteries, CPR images are less susceptible to artifacts in the large vessels of the extremities, where arteries course in relatively straight planes. Three-dimensional views are usually demonstrated with a lit projection and are helpful in demonstrating anatomic relationships, though they are problematic for demonstrating or grading stenoses (Fig. 17.6b).

Newer tools enable color-coding of vessels to bring attention to areas of plaque. Techniques are also available for characterization of atherosclerotic lesions with respect to attenuation values in order to classify lesions as fatty, fibrous, or calcified. These tools may be used adjunctively to the techniques described earlier, but have yet to be rigorously evaluated or validated.

Validation of Peripheral CT Angiography Advancement in CT technology has been rapid, with the recent advent of isotropic voxel imaging and multi-detector CT. The pace of technological advancement has surpassed the rate at which newer technologies are validated. For this reason, large multicenter studies and meta-analyses likely underestimate the accuracy of CT angiography as a tool.

17 CT Angiography of the Peripheral Arteries

303

Fig. 17.6 The thick MIP view (a) demonstrates the aortic stents. They are also well-seen on the volume rendered view (b) in this patient who is status post aortic stenting after repair for aortic coarctation and a bicuspid aortic valve

Large studies have, however, demonstrated several significant advantages of CT angiography in comparison to DSA, including a fourfold lower radiation dose and a much lower risk of complications. Moreover, studies have shown an excellent accuracy for the diagnosis of atherosclerotic disease as well as excellent correlation with DSA [20]. Diagnostic CT angiography performs comparably to DSA and favorably compares to duplex ultrasound and MR angiography for the evaluation in patients with chronic peripheral arterial disease or traumatic vascular injuries [21]. Compared to other non-invasive imaging modalities such as ultrasonography and MRA, CTA possesses several advantages. CTA reproducibility does not significantly depend on variability of technical skills as is oftentimes the limitation of ultrasonography. In patients with multilevel peripheral arterial disease, ultrasonography assessment has poor specificity to localize lesions, and is hindered by an impractical amount of time consumed in such extensive clinical evaluation. MRI angiography may have limitations in patients with stents, surgical clips, or other devices, and is problematic in patients with non-MR conditional cardiac devices. A study evaluating CT angiography with 64-row detector scanners for the detection of peripheral vascular disease evaluated 840 segments of the systemic arteries in 28 patients with lower extremity claudication. This study found an overall diagnostic accuracy of 98 % in the detection of lesions with a degree of stenosis of 50 % or higher. The sensitivity and specificity for

detecting stenoses by CT angiography were 99 % and 98 %, respectively [21]. Moreover, the use of advanced imaging tools, including 3-D reconstructions and multiplanar reformatted views, provide detailed visualization of stenotic lesions, normal vasculature, or previously revascularized lower extremity arteries along with nearby extravascular structures. Augmenting axial images with reformatted views has been shown to improve accuracy of interpretation [22]. Due to the speed and accessibility of imaging, CTA is also extremely useful in diagnosing acute limb ischemia and critical limb ischemia, helping clinicians to promptly and effectively formulate treatment plans. CTA also possesses advantages in depicting peripheral vascular aneurysms, providing clear, comprehensive images and precise dimensions along with delineating involvement of adjacent vessels and structures. Thus, CTA is a useful diagnostic and surveillance tool for aneurysm detection and follow-up.

Role of Peripheral CT Angiography for the Vascular Physician The most important application of CT angiography for the vascular interventional specialist is pre-procedure planning, including: selection of patients best treated with endovascular intervention versus open surgical procedures, identification of vascular access sites, pre-procedure selection of

J.E. Shriki et al.

304 Table 17.1 Advantages of CT angiography in patients with peripheral arterial disease prior to endovascular interventions 1. Selection of patients for endovascular versus open surgical revascularization. 2. Vascular access selection. 3. Lesion characterization (thrombus, degree of calcification, lesion length, vessel size). 4. Arterial vascular inflow and outflow. 5. Selection of interventional angiographic views and angulations. 6. Equipment selection based on lesion characteristics and vessel size (thrombolysis, sheaths, wires, balloons, stents, atherectomy, distal embolic protection). 7. Decreased contrast use. 8. Decreased radiation exposure. 9. Evaluation of extravascular arterial disease (popliteal entrapment, cystic adventitial disease, bony exostosis, thoracic outlet syndrome).

appropriate angiographic views, and pre-procedural lesion characterization (thrombus burden, dissection, calcification, tortuosity, etc.). CT angiography also provides valuable information for tailoring the most appropriate endovascular therapy, including: thrombolysis, laser, directional or orbital atherectomy, reentry device, distal embolic protection device, balloon angioplasty, self-expanding or balloon expandable stents, and covered stents (Table 17.1). Patients who undergo intervention for peripheral arterial disease have a higher incidence of vascular access site complications compared to patients who undergo percutaneous coronary intervention [23]. Patients with peripheral atherosclerotic disease commonly have a high burden of diffuse, often densely calcified, atherosclerotic plaques. As a result, vascular access selection is important to ensure safe and successful peripheral interventions. The atherosclerotic burden in some patients may prevent adequate hemostasis, which predisposes these patients to hemorrhagic complications at access sites. CT angiography offers an overall view of the arterial system and, therefore, may allow for identification of the most appropriate access site for peripheral interventions. In patients with severe, diffuse disease, alternative access sites or techniques (brachial, popliteal, antegrade, bypass grafts) may be utilized. CT angiography also helps in the decision to use distal embolic protection devices, especially in cases where there is heavy atherosclerotic burden, soft or unstable plaque, or thrombus. The choice of an appropriate device for the protection against distal embolization may be guided by vessel anatomy, tortuosity, and landing zone anatomy.

Normal Peripheral Arterial Anatomy and Variants Symptomatic manifestations of arterial diseases may appear in the distal extremities, but may also arise from disease which is proximal and remote to the site of symptoms. For example, non-healing ulcers in the toes as a result of isch-

emia may arise from stenosis as far proximal as the aorta. Since disease anywhere in the arterial tree may produce symptoms, knowledge of normal anatomy of the entire arterial tree is necessary in order to accurately interpret peripheral CT angiography.

Upper Extremities The normal upper extremity arterial supply begins with the subclavian arteries. The left subclavian artery typically arises directly from the aortic arch. The right subclavian artery most commonly arises from the brachiocephalic (innominate) artery, which typically gives off a right common carotid artery as well a right subclavian artery. In 15 % of patients, the innominate artery also gives off the left common carotid artery, a variant described as a bovine arch. In these patients, the left common carotid artery commonly arises as the first vessel off of the bovine innominate, although it may arise more cranially as a trifurcation vessel of the innominate artery [24]. Other commonly encountered variants of aberrant origination of the subclavian artery exist. An aberrant right subclavian artery may either arise from a left sided aortic arch, as the last major vessel from the arch (left arch with aberrant right subclavian artery) (Fig. 17.7). In the case of a right aortic arch, the left subclavian artery may arise as the last major vessel from the arch (right arch with aberrant left subclavian artery). In either case, the aberrant subclavian artery usually takes a course posterior to the esophagus and may produce dysphagia, which is commonly referred to as dysphagia lusoria. A double aortic arch may also cause dysphagia lusoria [25]. In addition, an aberrant subclavian artery may arise from a dilated trunk, termed a diverticulum of Kommerel. This is a true aneurysm that likely results from an embryological remnant of a separate, incompletely formed aortic arch. Both the double aortic arch and the right arch with an aberrant left subclavian artery represent vascular rings. In the latter case, the ring is completed by the ligamentum arteriosum. Clinically significant atherosclerotic occlusive complication rates resulting from aberrant subclavian arteries are likely similar to rates of atherosclerotic complications observed in normal arteries, although aberrant subclavian arteries are more prone to aneurysmal degeneration. Anatomically, the subclavian artery is divided into proximal, middle, and distal portions. The proximal portion of the subclavian artery is defined as the portion medial to the anterior scalene muscle. The mid portion of the subclavian artery is located posterior to the anterior scalene muscle and usually contains the most cranial portion of the subclavian arch. The distal portion of the subclavian artery lies lateral to the lateral border of the anterior scalene muscle and ends at the lateral border of the first rib. At this point the subclavian artery changes name to become the axillary artery. The vertebral artery is usually the first vessel that arises from the subclavian artery and most commonly arises from the

17 CT Angiography of the Peripheral Arteries

305

Fig. 17.7 On this CT scan of the chest performed to evaluate for an etiology of shortness of breath, a right-sided aortic arch with an aberrant left subclavian artery (white arrow) is incidentally noted. This is shown on the transverse view (a) and volume rendered (b) view

first portion of the subclavian artery, usually within 1.2–2.5 cm of the vessel origin. The other vessels include the internal mammary artery (Fig. 17.8), thyrocervical trunk, and costocervical trunk. These vessels also most commonly arise from the first portion of the subclavian artery and are usually clustered near the medial border of the anterior scalene muscle. At the lateral border of the first rib, the subclavian artery transitions to become the axillary artery, which proceeds to predominantly supply arterial blood flow to the upper chest wall and the proximal portion of the upper extremity. In the case of axillary artery occlusion proximal to the origin of the subscapular artery, collateral flow may be provided through chest wall and scapular collaterals. By definition, the axillary artery ends at the lateral border of the teres major, where it changes name to become the brachial artery. The axillary artery is surrounded by the brachial plexus. The brachial artery is the main vessel to the upper extremity. Most commonly, the brachial artery gives off a profunda branch in the upper portion of the upper extremity. Below the elbow, the brachial artery usually trifurcates into a radial artery laterally and a common trunk that gives off an interosseous artery and an ulnar artery medially. In 15 % of patients, the brachial artery gives off the radial artery proximal to the elbow as it courses in the upper arm. The radial or ulnar artery may rarely arise aberrantly from the axillary artery. There is a close relationship with the median nerve which normally runs just medial to the brachial artery throughout the upper arm. The median nerve may overlie the brachial artery rendering it vulnerable to injury during brachial artery access.

Lower Extremity The aortic bifurcation most commonly occurs at the level of the L4 vertebral body, although some patients may have an unusually high aortic bifurcation as a normal variant. The common iliac arteries are usually 4–5 cm in length, although the right common iliac artery is usually slightly longer than the left. The common iliac arteries course medial to the psoas muscles and beneath the ureters to the inferior pelvic brim before bifurcating into external and internal branches. The common iliac artery may give rise to an iliolumbar trunk, which can be a source of endoleak in patients who have undergone aortic aneurysm repair. Accessory renal arteries may rarely arise from the common iliac arteries, especially when a pelvic or ptotic kidney is present. The internal iliac artery runs posteriorly from the common iliac artery, and subsequently gives off anterior and posterior divisions. The main branch of the posterior division is the superior gluteal artery, which exits the sciatic foramen. The posterior division may also give rise to an iliolumbar artery. The anterior division gives off several important branches including the internal pudendal artery, and the uterine artery in women. The external iliac artery courses more anteriorly in comparison to the internal iliac artery. Below the inguinal ligament, it changes name to become the common femoral artery. Vascular landmarks for the inguinal ligament are the origins of the deep circumflex iliac artery and the inferior extent of the inferior epigastric artery, which also mark the delineation between external iliac artery and

306

a

J.E. Shriki et al.

b

Fig. 17.8 CT angiography is useful in demonstrating the internal mammary arteries in patients in whom aortocoronary bypass is planned. The course of the left internal mammary artery (white arrows) is demonstrated on the curved plane reformatted view (a) and also on the

volume rendered view (b). In evaluating the subclavian artery and its branches, injection of contrast should be made via the contralateral extremity in order to ensure that streaking from dense venous contrast does not occur

common femoral artery. The external iliac artery gives off other small branches, such as small muscular branches and the cremasteric artery, which runs in the spermatic cord. The common femoral artery, which begins below the inguinal ligament, is a short vessel which usually has a length of 4 cm and gives off the superficial femoral artery and deep femoral artery at approximately the level of the lesser trochanter of the femur. Most commonly, there is also a slightly more lateral branch given off at the same level, which is the circumflex femoral artery. The term “superficial femoral artery” is still used commonly by physicians, whereas anatomists favor the name “femoral artery”. In regard to femoral venous anatomy, there has been some shift in nomenclature among physicians in order to avoid confusion when thrombi of the vein are reported. Drop of the descriptor “superficial”

from the femoral vein can avoid confusion as the vein is part of the deep venous system. The femoral artery courses anteromedially in the thigh. In the middle third of the thigh, the femoral artery enters the adductor canal or eponymously, Hunter’s canal. This is a frequent site of atherosclerotic disease. At the junction of the middle and lower third of the thigh, the femoral artery exits the adductor canal and changes name to the popliteal artery. The popliteal artery is a common site of several unique diseases including cystic adventitial disease and popliteal artery entrapment syndrome. Knee dislocation injuries may damage the popliteal artery. At approximately the level of the knee, the popliteal artery gives off medial and lateral geniculate branches. These branches may serve as important collaterals for reconstitution of the popliteal artery from the

17 CT Angiography of the Peripheral Arteries

profunda using geniculate collateral pathways in the setting of femoral artery occlusion. The popliteal artery continues behind the knee and gives off the anterior tibial artery. In most patients, the anterior tibial artery gives off the dorsalis pedis artery, which courses along the dorsal aspect of the foot. The other main branch of the popliteal artery is the tibioperoneal trunk. This divides into the posterior tibial artery and peroneal artery. Variations in this conventional anatomy occur approximately 10 % of the time, with the most common variant being a high takeoff of the anterior tibial at or above the level of the knee joint. A true trifurcation followed by a hypoplastic or absent posterior tibial artery are the next most common variants, while high origin of the posterior tibial artery is rare. The peroneal artery runs in the deep compartment of the lower portion of the lower extremity and usually terminates above the ankle in collateral branches to the posterior tibial and dorsalis pedis arteries. The posterior tibial artery runs posterior to the medial maleolus of the ankle and frequently can be palpated at this point. The plantar arch is an arcade of vessels which may be primarily served by either the dorsalis pedis artery or the posterior tibial artery. The main vessel supplying the plantar arch should be noted and included in CT angiography reports.

a

Fig. 17.9 (a, b) An unusual variant is shown in which there is congenital absence of the external iliac artery. The internal iliac artery (white arrow) in this case takes a course posteriorly to give off the

307

CT has an advantage in comparison to conventional angiography in demonstrating 3-D vascular anatomy and variants in the context of muscular and osseous anatomy. Variants which contribute to clinically significant arterial disease are rare and some variants may not cause significant vascular pathology. Rarely, for example, the external iliac artery may be absent, and the common femoral artery arises from the internal iliac artery (Fig. 17.9). This common variant would not be expected to cause clinically significant manifestations of arterial disease [26]. Rarely, the main lower extremity artery may arise from the internal iliac artery, and courses posteriorly to the ischial tuberosity. This variant is known as a persistent sciatic artery. The anomalous course of the artery along the ischial tuberosity can result in premature atherosclerotic disease (Fig. 17.10) and also in formation of aneurysms (Fig. 17.11). Typically, occlusion or aneurysm formation occurs where the artery courses behind the ischial tuberosity. Vascular pathology is thought to occur due to repetitive underlying trauma due to impact of the ischial tuberosity onto the artery [27]. For variants where abnormal muscular anatomy may contribute to pathology, the arterial tree may be better imaged with MRI. In general imaging of the muscular structures of the extremities, MRI has advantages relative to CT angiography,

b

anterior and posterior divisions, before continuing anteriorly to give off the common femoral artery

308

J.E. Shriki et al.

a

b

Fig. 17.10 (a, b) Images of the right lower extremity are shown in a patient with bilateral persistent sciatic arteries. Note that the persistent sciatic artery (white arrows) is occluded at the level of the ischial tuberosity (white arrowhead). There is also a second long segment of

probable occlusion of the persistent sciatic artery in the thigh. There is, however, reconstitution of the vessel via large profunda collaterals after both occlusions

including better resolution of soft tissue structures such as musculature and joints. Non-contrast MR angiography techniques also permit imaging with the extremity in different positions, without the use of ionizing radiation.

as their initial manifestation. The presence of peripheral arterial disease significantly contributes to worsened morbidity and mortality, likely because it is a marker of systemic atherosclerotic disease burden. Patients with peripheral arterial disease have a four to fivefold increase in risk of myocardial infarction or stroke [29, 30]. Peripheral arterial disease is a common condition, occurring in 10–25 % of patients over the age of 55. The incidence increases with age at a rate of 0.3 % per year in men aged 40–55 and at a rate of 1 % per year in men over the age of 75. Up to 70–80 % of affected individuals are symptomatic, although only a minority of patients will eventually require revascularization. Twenty five percent of patients with peripheral arterial disease will require some medical or surgical treatment. Because of their increased morbidity and mortality, all patients with peripheral arterial disease should

Abnormalities and Diseases of the Peripheral Arteries Atherosclerotic Disease Atherosclerosis is by far the most common disease of the peripheral arterial tree [28]. At times, patients may present with concomitant cerebrovascular disease and coronary atherosclerotic disease, although some patients with atherosclerotic disease may present with peripheral ischemia

17 CT Angiography of the Peripheral Arteries

a

309

b

c

Fig. 17.11 (Same patient as in Fig. 16.9). (a–c) Images of the left lower extremity are shown in a patient with bilateral persistent sciatic arteries (white arrows). Note the presence of a large aneurysm extending

just above the ischial tuberosity (white arrowhead). Note also that the vessel is less well opacified distal to the aneurysm due to stagnant flow

have aggressive control of their atherosclerotic risk factors; however, only about 25 % of patients are actually treated [31, 32]. Peripheral CT angiography can demonstrate atherosclerosis at a very early stage in the disease process. Low attenuation plaques, likely related to an early phase of plaque evolution, may be seen [33]. Plaques may be calcified or non-calcified, and may not cause stenosis until very late in the disease course (Fig. 17.12). Densely calcified atherosclerotic plaques in the peripheral arterial tree may somewhat degrade CT angiography image quality, although this is less of a concern when imaging the extremities compared to the coronary arteries due to the larger caliber of vessels and the lower potential for motion and other artifacts. Atherosclerotic plaques may be present throughout the vascular tree. Typically aortic atherosclerotic disease begins in the infrarenal aorta and becomes more severe closer to the aortic bifurcation. A rarer variant in some patients with atherosclerotic disease is the development of arborified, endoaortic calcified plaques, which predominantly protrude into the lumen, and are usually most pronounced in the juxtamesenteric and juxtarenal aorta. This variant has been termed a “coral reef” aorta (Fig. 17.13) [34]. Recognition of this variant of atherosclerotic disease is important since patients with endoaortic calcific proliferation are at higher risk for postcatheterization embolic phenomenon. It has been suggested

that patients with a ‘coral reef aorta’ should not undergo endovascular interventions which necessitate crossing of the juxtamesenteric aorta should be avoided in patients with a “coral reef aorta” [35, 36]. In the peripheral tree, atherosclerotic disease may be multifocal and usually consists of mixed attenuation plaques. CT angiography is useful in demonstrating stenoses of 50 % or greater, which may contribute to patient symptoms. Specific features of each plaque that should be described include the location of the lesion, degree of stenosis, and length of the plaque. When CT angiography is used for plaque characterization, descriptors for plaque attenuation may be added, with reporting of plaques as calcified, non-calcified, or mixed plaque. Further evaluation of lesions with stenoses is commonly pursued with catheterization for measurement of pressure gradients. Specific criteria for intervention have also been delineated, based on the degree of patient symptoms [37]. Atherosclerotic disease may cause a number of symptoms depending on the site of involvement. Several syndromes have been characterized based on the distribution of atherosclerotic disease. For example, subclavian steal syndrome results from proximal stenosis in the subclavian artery (Fig. 17.14). As a result of stenosis, the distal subclavian artery may receive collateral flow from the vertebral arteries. Reversal of flow through the ipsilateral vertebral artery commonly ensues. Because of the relatively rich brain collateral

310

a

J.E. Shriki et al.

b

Fig. 17.12 (a, b) Multifocal atherosclerosis is demonstrated on these volume rendered views of the lower extremities. The femur, tibia, and fibula have been subtracted from the field of view in order to better demonstrate the arterial anatomy

system, only a small percentage of patients with this reversal of flow will present with symptoms related to vertebrobasilar insufficiency. These symptoms are commonly worsened during exercise of the upper extremity, which results in increased flow to the extremity and worsened steal from the cerebrovascular circulation. Leriche syndrome is a constellation of symptoms which results from aortic and bilateral iliac artery disease, including gluteal and lower extremity claudication, penile impotence, and lower extremity atrophy. Similar to assessment of aortic aneurysms, duplex ultrasound is the most cost effective strategy for surveillance of popliteal or femoral aneurysms. CT angiography, however, is also favored over other imaging modalities for accurate sizing of the proximal and distal arterial landing zones prior to endovascular peripheral aneurysm repair. Evaluation of thrombus and patency of runoff vessels is also easily accomplished by preoperative CT angiography.

Grafts and Stents in the Arterial Tree In addition to being a non-invasive modality with excellent spatial resolution, CT angiography has several other advantages in the evaluation of the treated vascular system. In comparison to MRI and MR angiography, CT angiography is advantageous for visualization of stents. On MRI, stents may be visualized only as artifacts and the internal lumen may be non-visualized due to susceptibility effects. Even when the internal lumen is visualized, the stented segment generally is incompletely evaluated by MR angiography. Other metallic structures including surgical clips may also induce artifacts on MRI, including signal void and failure of fat saturation, whereas artifacts from surgical devices are generally less significant on CT. Dual source or dual energy CT further potentiates visualization of high attenuation, metallic structures with

17 CT Angiography of the Peripheral Arteries

a

311

b

c

Fig. 17.13 Views from a CT angiogram of the abdomen are shown with transverse (a, b) and sagittal (c) images shown. A “coral reef” aorta is present with dense endoaortic calcific proliferation. Densely

calcified, endoluminal, arborified plaques are present (white arrows) in the juxtamesenteric aorta

less significant obscuration of adjacent anatomy due to minimizing of streaking. Stents in peripheral arterial structures are typically well-seen using thick MIP images (Fig. 17.15). This allows visualization of stent struts and exclusion of strut fractures. Stents are easily depicted as high attenuation structures. Stents are frequently well-evaluated on postcontrast and non-contrast images. In the short axis view, stent struts are frequently seen as regularly spaced, hyperattenuating foci at the rim of the artery, commonly in a hexagonal array (Fig. 17.16). Because of the relatively

high attenuation of metallic stents, and because of the phenomenon of “blooming” on CT, a very bright stent may appear to be outside the confines of the wall of a vessel. The limitations of stent depiction on coronary CT are less significant in evaluation of the peripheral arterial tree due to the larger internal diameter of stents commonly employed in the peripheral vessels and also due to the absence of motion and other artifacts that can limit the evaluation of stented coronary arteries. In-stent restenosis in the peripheral vasculature is usually easily evaluated using CT angiography.

312 Fig. 17.14 Images from a CT angiogram are shown in a patient with known coronary artery disease and concomitant symptoms of subclavian steal. The oblique sagittal (a) and volume rendered (b) views show a focal, shelf-like area of narrowing near the origin of the left subclavian artery (arrow, (a) and (b)). In this case, identification of this stenosis was useful as an explanation of the patient’s symptoms. Preoperative identification of subclavian stenosis is also important in patients in whom aortocoronary bypass is planned, as this condition may impede optimal flow through the internal mammary artery

a

J.E. Shriki et al.

a

b

b

Fig. 17.15 Stents are well-depicted on CT angiography. In this case, the stent is seen on the volume rendered view (yellow arrow, a) and also on the orthogonal, curved plane reformatted views (white arrows, b)

17 CT Angiography of the Peripheral Arteries

a

313

b

c

Fig. 17.16 A stent is the left common iliac artery is shown. The stent is present on the volume rendered view (a) and also on the curved plane reformatted views (b). Note that, in the short axis of the vessel (c), the stent is seen as a hexagonal array of hyperattenuating struts

a

b

d

c

Fig. 17.17 Bilateral, aortofemoral bypass grafts are present (white arrows) and are seen as unusually smooth appearing structures connecting portions of the vascular tree. The occluded, native vessels are visualized on the transverse view (b, white arrowheads), but are not visualized on the volume rendered view (a) since the native arteries are

not opacified. Enlargement and irregularity may be present at anastomotic sites as shown in this transverse CT image taken at the level of the patient’s anastomoses (white, open arrowheads, c). Note that the patient also has aortic and celiac stents (black arrows, d)

Graft material is also well evaluated on CT. Bypass grafts are commonly recognized as long, smooth, branchless tubes connected to the native vasculature on 3-D colored, lit projections (Fig. 17.17). On axial views, the excluded, unopacified, native vessels are frequently visible. The connections between graft material and native vessel lumen may be

enlarged and irregular, as a result of the patch angioplasty frequently performed at anastomosis sites. Grafts commonly are comprised of either interposed veins, Dacron, or expanded polytetrafluoroethylene (PTFE). In some cases, where increased torsional effects are anticipated and may compromise grafts, reinforced graft material is commonly

314

a

J.E. Shriki et al.

b

Fig. 17.18 A bifemoral bypass graft (white arrows) is evident with a typical, corrugated appearance, which is well seen on the volume rendered view (a) and the curved planar reformatted view (b)

employed. The rings of such grafts are typically visible as corrugated on CT angiography (Fig. 17.18).

Trauma CT angiography as a modality has multiple features that make it ideal for imaging of the arterial tree in the setting of trauma. First, intimal flaps and abnormalities of the wall of the artery are better depicted by CT angiography compared to MR angiography, and may be better seen on CT than on ultrasound, especially within the bony pelvis where bowel gas and patient body habitus may limit duplex evaluation. CT angiography is also useful in demonstrating the entire arterial tree in a less time-intensive fashion than ultrasound or MR angiography. Concomitant post-traumatic deformities to the muscles and bones may also be simultaneously demonstrated on CT (Fig. 17.19). CT signs of arterial injury include contrast extravasation, vessel non-opacification, abrupt vessel occlusion, focal vessel narrowing/spasm, pseudoaneurysm, intimal flap, or arteriovenous fistula. Traumatic injury to vessels may ensue after blunt or penetrating trauma and may be seen in association with fractures which displace vessels [16].

Fibromuscular Dysplasia Although fibromuscular dysplasia is a common cause of stenosis in the renal or carotid arteries, it is less commonly encountered elsewhere in the peripheral arterial tree. When

involving the peripheral arteries, fibromuscular dysplasia most commonly occurs in the external iliac artery, which is the third most common site of fibromuscular dysplasia in the body. As in other parts of the body, the classification system for fibromuscular dysplasia is based on the layer of the artery involved, with medial fibroplasia being the most common form. The most typical appearance of fibromuscular dysplasia is apparent beading of the vessel and is due to several, closely approximated weblike areas of narrowing with intervening outpouchings of the vessel from post-stenotic dilatation (Fig. 17.20) [38]. Other forms of fibromuscular dysplasia may have a variety of appearances [39]. Conventional angiography may have an advantage in demonstrating this entity compared with CT, due to the inherently higher spatial resolution of conventional angiographic images.

Other Diseases of the Systemic Arteries Cystic adventitial disease is a rare entity, which may affect any artery adjacent to a joint and presents as a smooth narrowing without atherosclerotic disease. The narrowing is accompanied by cystic structures along the course of the artery. The most common artery affected is the popliteal. MRI is the preferred modality for depicting the cysts which occur along the vessel, although low attenuation cysts are commonly observed on CT [40, 41]. Popliteal artery entrapment syndrome can occur due to a number of abnormalities in the relationship between the popliteal artery and the muscles of the popliteal space. The most common abnormal muscle in this case is the medial

17 CT Angiography of the Peripheral Arteries

315

a

b

Fig. 17.19 CT angiography is useful in the setting of trauma. A surface-rendered view (a) shows the deformity in the outer contour of the extremity. CT angiography simultaneously demonstrates osseous structures, demonstrating a dislocation at the knee (b, c). The osseous

a

Fig. 17.20 Fibromuscular dysplasia is shown in the external iliac artery, which is the third most common site for this entity, following the internal carotid and renal arteries. The classic, beaded appearance of the

structures may be subtracted, however, in order to better demonstrate the underlying arterial anatomy (d). In this case, resultant occlusion of the popliteal artery is also present (white arrow, d)

b

external iliac artery (arrow) is demonstrated on a reformatted view from the patient’s abdominal CT (a), but is more clearly demonstrated on the conventional angiography (b), due to the higher spatial resolution

316

head of the gastrocnemius, although a number of abnormal relationships have been described. This syndrome usually causes some degree of fixed narrowing of the popliteal artery, although there is commonly a dynamic component of narrowing, usually during plantar flexion or dorsiflexion. Repetitive trauma to the artery as a result of the abnormal relationship to the muscle may cause aneurysmal dilatation, thrombosis, or thromboembolism. MRI is useful in demonstrating popliteal artery entrapment syndrome, where an abnormal muscular slip courses medial to the popliteal artery. In this case, the lower extremity may need to be imaged in several positions including dorsiflexion and plantarflexion [40]. This is also more easily performed with MR angiography, since MR angiography is less sensitive to optimal vascular opacification and images can be obtained at different time-points. Also, as non-contrast means of performing MRA become more robust, some vascular pathology may be imaged without the administration of contrast. Since the common femoral artery is a common site of vascular access, it is subject to iatrogenic complications including chiefly pseudoaneurysm and arteriovenous fistula formation. Because of the focal nature of these complications, and because the portion of the artery involved is frequently very superficial, ultrasound with Doppler is usually an adequate modality for the diagnosis and follow-up of iatrogenic femoral artery complications. On the other hand, when a deep or retroperitoneal hematoma is suspected, CT may be a more robust technique than ultrasound.

Other Modalities for Imaging the Peripheral Arteries Advancements in imaging of the peripheral arteries have occurred in virtually every modality. As a result, the decision between modalities is more complex. Physical exam and ankle-brachial index measurement is an adequate means of making an initial diagnosis of peripheral arterial disease [37]. Further evaluation with ultrasound is also useful in demonstrating and localizing atherosclerotic disease. Complete evaluation of the entire extremity with ultrasound is, however, very time-intensive and detection of disease is technologist dependent. Detection and measurement of stenoses with ultrasound is also dependent on technical factors, such as the angle of insonation employed. Evaluation of the pelvic vasculature by ultrasound is much more difficult, and portions of the vasculature may not be easily demonstrated with ultrasound due to overlying bowel gas and osseous structures. In very obese patients, ultrasound may be significantly limited. Heavy or circumferential calcification, such as that found within patients with diabetes or renal insufficiency also significantly limits ultrasound. Determination of severity of disease by ultrasound also has difficulty in determining the severity of disease in arterial segments distal to a high-grade stenosis.

J.E. Shriki et al.

MRI and MR angiography have several advantages in patients, including the ability to perform imaging without contrast. Non-contrast MR angiography techniques have advanced dramatically, although there is still considerable variability between institutions and MR technology. Clinically useful imaging of tibial and pedal vessels using non-contrast MR is generally not possible except in highly specialized centers. Because calcium does not interfere with contrastenhanced MR angiography, evaluation of tibial vessels with MRA, when a separate tibial imaging bolus is used instead of a bolus-chase technique, is frequently superior to CTA in patients with critical limb ischemia. In particular, the adequacy of MR sequences for imaging the arterial tree are dependent on the scanner, sequences, and vendor-specific techniques used. MR angiography has significant limitations in evaluating the post-surgical arterial tree, due to artifacts such as failure of fat saturation and susceptibility artifacts due to surgical clips, stents, or other foreign material. MR angiography is also contra-indicated in patients with non-MR conditional pacemakers or ICDs. Likewise, certain stents and stent-grafts are MR conditional, such that patients with these implants cannot undergo MR in 3 T machines. Patients with claustrophobia or significant back pain may not tolerate lying still for the hour-long exam. Because of these limitations MR is contraindicated in approximately 30 % of patients. In the past, MR has been preferable in patients with renal disease due to relatively lower nephrotoxicity of gadolinium, compared to iodinated contrast media. However, the recent recognition of nephrogenic systemic fibrosis as a complication of gadolinium administration has decreased the utility of MR angiography in patients with chronic, severe renal disease [42]. Gadolinium should generally not be given to patients with a creatinine clearance of 30 ccs per minute or less. In patients who are already dialysis-dependent, iodinated contrast may be a better choice. CT has an advantage to MR angiography in superior spatial resolution and depiction of smaller vessels. Evaluation of the patency of circumferentially calcified tibial vessels remains challenging for CTA, however, and is one of the few circumstances where DSA may be required.

Radiation Dose in Peripheral CT Angiography The radiation dose in CT angiography remains high and is increasingly a consideration in most CT applications. Concerns of radiation are somewhat mitigated by the fact that the extremities contain less radiosensitive tissues. When imaging the extremities, breast and abdominal shielding can easily be employed with no compromise to image quality. Shielding significantly decreases scatter and is under-utilized in patients undergoing CT in general, including peripheral CT angiography.

17 CT Angiography of the Peripheral Arteries

The radiation dose for conventional angiography is, however, much higher than for CT angiography [43]. This is in contradistinction to radiation doses in the heart, where catheterization results in lower radiation doses compared to CT. One study found that for a 16-slice CT scanner, the average radiation dose for a peripheral CT angiogram was 3.0 mSv in men, whereas the radiation dose for a conventional angiogram had an average of 11.0 mSv. Other studies have shown similar results, with CT angiography generally found to have a fourfold lower radiation dose in comparison with peripheral angiography [44]. Although peripheral CT angiography has a relatively low radiation dose and relatively less radiosensitive tissues are exposed, the risks of radiation should not be taken lightly.

References 1. Chiro GD, Brooks RA, Kessler RM, et al. Tissue signatures with dual-energy computed tomography. Radiology. 1979;131:521–3. 2. Flohr TG, McCollough CH, Bruder H, et al. First performance evaluation of a dual-source CT (DSCT system). Eur Radiol. 2005;16(2):256–68. 3. Kau T, Eicher W, Reiterer C, et al. Dual-energy CT angiography in peripheral arterial occlusive disease-accuracy of maximum intensity projections in clinical routine and subgroup analysis. Eur Radiol. 2011;21(8):1677–86. 4. Rubin GD, Schmidt AJ, Logan LJ, Sofilos MC. Multi-detector row CT angiography of lower extremity arterial inflow and runoff: initial experience. Radiology. 2001;221:146–58. 5. Ofer A, Nitecki SS, Linn S, et al. Multidetector CT angiography of peripheral vascular disease: a prospective comparison with intraarterial digital subtraction angiography. Am J Roentgenol. 2003;180:719–24. 6. Jakobs TF, Wintersperger BJ, Becker CR. MDCT-imaging of peripheral arterial disease. Semin Ultrasound CT MR. 2004; 25:145–55. 7. Johnson PT, Pannu HK, Fishman EK. IV contrast infusion for coronary artery CT angiography: literature review and results of a nationwide survey. Am J Roentgenol. 2009;192:W214–21. 8. Bae KT, Seeck BA, Hildebolt CF, et al. Contrast enhancement in cardiovascular MDCT: effect of body weight, height, body surface area, body mass index, and obesity. Am J Roentgenol. 2008; 190:777–84. 9. Fleischmann D. Aorto-popliteal bolus transit time in peripheral CT angiography: can fast acquisition outrun the bolus? Eur Radiol. 2003;13:S268. 10. Heuschmid M, Krieger A, Beierlein W, et al. Assessment of peripheral arterial occlusive disease: comparison of multislice-CT angiography (MS-CTA) and intraarterial digital subtraction angiography (IA-DSA). Eur J Med Res. 2003;8:389–96. 11. Cademartiri F, Mollet NR, Van der Lugt A, et al. Intravenous contrast material administration at helical 16–detector row CT coronary angiography: effect of iodine concentration on vascular attenuation. Radiology. 2005;236:661–5. 12. Orlandini FA, Boini S, Iochum-Duchamps S, et al. Assessment of the use of a saline chaser to reduce the volume of contrast medium in abdominal CT. Am J Roentgenol. 2006;187:511–5. 13. Cademartiri F, Mollet NR, Van der Lugt A, et al. Non-invasive 16-row multislice CT coronary angiography: usefulness of saline chaser. Eur Radiol Vol. 2004;14(2):178–83.

317 14. Becker CR, Wintersperger B, Jakobs TF. Multi-detector-row CT angiography of peripheral arteries. Semin Ultrasound CT MR. 2003;24:268–79. 15. Fleischmann D. Use of high concentration contrast media: principles and rationale—vascular district. Eur J Radiol. 2003;45:S88–93. 16. Miller-Thomas MM, West OC, Cohen AM. Diagnosing traumatic arterial injury in the extremities with CT angiography: pearls and pitfalls. Radiographics. 2005;25:S133–42. 17. LawrenceJ A, Kim D, Kent KC, et al. Lower extremity spiral CT angiography versus catheter angiography. Radiology. 1995; 194:903–8. 18. Rieker O, Duber C, Schmiedt W, et al. Prospective comparison of CT angiography of the legs with intraarterial digital subtraction angiography. AJR Am J Roentgenol. 1996;166:269–76. 19. Otah KE, Takase K, Igarashi K, et al. MDCT compared with digital subtraction angiography for assessment of lower extremity arterial occlusive disease: importance of reviewing cross-sectional images. AJR Am J Roentgenol. 2004;182:201–9. 20. Willman JK, Baumert T, Chandler P. Aortoiliac and lower extremity assessed with 16 detector row CT angiography respective comparison with distal subtraction angiography. Radiology. 2005;236: 1083–93. 21. Shareghi S, Gopal A, Gul K, et al. Diagnostic accuracy of 64 multidetector computed tomographic angiography in peripheral vascular disease. Catheter Cardiovasc Interv. 2010;75:23–31. 22. Fishman EK, Ney DR, Heath DG, et al. Volume rendering versus maximum intensity projection in CT angiography: what works best, when, and why. Radiographics. 2006;26:905–22. 23. Shammas MW, Lemke JH, Dipple EG. In-hospital complications of peripheral vascular interventions using fractionated heparin as the primary anticoagulant. J Invasive Cardiol. 2003;15:242–6. 24. Lippert H, Pabst R. Aortic arch. In: Arterial variations in man: classification and frequency. Munich: JF Bergmann-Verlag; 1985. p. 3–10. 25. Janssen M, Baggen MGA, Veen HF, et al. Dysphagia lusoria: clinical aspects, manometric findings, diagnosis, and therapy. Am J Gastroenterol. 2000;95:1411–6. 26. Koyama T, Tadanori T, Kitanaka Y, Katagiri K, et al. Congenital anomaly of the external iliac artery: a case report. J Vasc Surg. 2003;37(3):683–5. 27. Brantley SK, Rigdon EE, Raju S. Persistent sciatic artery: embryology, pathology, and treatment. J Vasc Surg. 1993;18(2):242–8. 28. Mesurolle B, Qanadli SD, El Hajjam M, Goeau-Brissonniere OA, Mignon F, Lacombe P. Occlusive arterial disease of abdominal aorta and lower extremities: comparison of helical CT angiography with transcatheter angiography. Clin Imaging. 2004;28:252–60. 29. Ness J, Aronow WS. Prevalence of coexistence of coronary artery disease, ischemic stroke, and peripheral arterial disease in older persons, mean age 80 years, in an academic hospital-based geriatrics practice. Brief reports. J Am Geriatr Soc. 1999;47(10):1255–6. 30. Newman AB, Shemanski L, Manolio TA, et al. Ankle-arm index as a predictor of cardiovascular disease and mortality in the cardiovascular health study. Arterioscler Thromb Vasc Biol. 1999; 19:538–45. 31. “Peripheral arterial disease prevention and prevalence”. Peripheral arterial disease. 2007. http://www.3-rx.com/ab/more/peripheralarterial-disease-prevention-and-prevalence/. Retrieved on 03 Sept 2009. 32. Sharrett AR. “Peripheral arterial disease prevalence”. Peripheral arterial disease. 2007. http://www.health.am/vein/more/peripheralarterial-disease-prevalence/. Retrieved on 03 Sept 2009. 33. Cordeiro M, Lima J. Atherosclerotic plaque characterization by multidetector row computed tomography angiography. J Am Coll Cardiol. 2006;47(8):C40–7. 34. Rosenberg GD, Killewich LA. Case report: blue toe syndrome from a “coral reef” aorta. Ann Vasc Surg. 1995;9(6):561–4.

318 35. Levien LJ, Veller MG. Popliteal artery entrapment syndrome: more common than previously recognized. J Vasc Surg. 1999; 30(4):587–98. 36. Qvarfordt PG, Reilly LM, Sedwitz MM, Ehrenfeld WK, Stoney RJ. “Coral reef” atherosclerosis of the suprarenal aorta: a unique clinical entity. J Vasc Surg. 1984;1(6):903–9. 37. Hirsch AT, et al. ACC/AHA guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic). J Vasc Interv Radiol. 2006;17(9):1383–98. 38. Walter JF, Stanley JC, Mehigan JT, et al. External iliac artery fibrodysplasia. Am J Roentgenol. 1978;31(1):125–8. 39. Sauer L, Reilly LM, Goldstone J, et al. Clinical spectrum of symptomatic external iliac fibromuscular dysplasia. J Vasc Surg. 1990;12(4):488–95. Discussion 495–6. 40. Elias DA, White LM, Rubenstein JD, et al. Pictorial essay, clinical evaluation and MR imaging features of popliteal artery entrapment

J.E. Shriki et al.

41.

42.

43.

44.

and cystic adventitial disease. Am J Roentgenol. 2003;180: 627–32. Tsolakis CS, Walvatne CS. Cystic adventitial disease of the popliteal artery: diagnosis and treatment I.A. Eur J Vasc Endovasc Surg. 1998;15(3):188–94. Sadowski EA, Bennett LK, Chan MR, et al. Nephrogenic systemic fibrosis: risk factors and incidence estimation. Radiology. 2007;243:148–57. Kocinaja D, Cioppaa AA, Ambrosinia GT, et al. Radiation dose exposure during cardiac and peripheral arteries catheterization. Int J Cardiol. 2006;113(2):283–4. Martin ML, Tay KH, Borys F, et al. Multidetector CT angiography of the aortoiliac system and lower extremities: a prospective comparison with digital subtraction angiography. Am J Roentgenol. 2003;180:1085–91.

Aortic, Renal, Mesenteric and Carotid CT Angiography

18

Anas Alani and Matthew J. Budoff

Abstract

Computed tomography angiography has an increasing role in vascular imaging of the aorta, renal, mesenteric, and carotid arteries. There has been tremendous improvement in computed tomography technology that has made such images the preferred choice for diagnosing various acute and chronic vascular diseases and replacing non-invasive and invasive tests. Keywords

Computed Tomography Angiography • Aorta CT Angiography • Renal CT Angiography • Mesenteric CT Angiography • Carotid CT Angiography • Vascular CT Angiography • Aortic Dissection • CT Angiography Acquisition and Protocol

Introduction Computed tomographic angiography (CTA) of vascular beds is significantly easier to perform and interpret than coronary studies. There is no cardiac motion to contend with, so gating is most often not necessary. The exception is the ascending aorta, where pseudodissections (an appearance of a dissection caused by motion of the aortic root – Fig. 18.1) have plagued earlier studies with single-slice computed tomography (CT) due to motion artifacts [1]. Most of the large vessels of interest (the carotid, renal, and mesenteric arteries) have significantly larger diameters than

A. Alani, MD Department of Medicine, University of Florida – Gainesville, Gainsville, FL, USA Los Angeles Biomedical Research Institute at Harbor-UCLA, 1124 W Carson Street, Torrance, CA 90502, USA e-mail: [email protected] M.J. Budoff, MD (*) David Geffen School of Medicine at UCLA, Los Angeles Biomedical Research Institute, Torrance, CA USA e-mail: [email protected]

coronary arteries, as well as less tortuous courses. The renal and carotid arteries are usually straight structures, so reconstructions are significantly less complicated than coronary imaging. Also, due to the increased speed of newer systems (electron beam tomography (EBT) and 16+ row multidetector computed tomography (MDCT)), venous enhancement is less common, so it is easier to see the arteries without superimposed contrast-filled structures (venous contamination). This is another reason why CT is most often superior to magnetic resonance imaging (MRI) in these vascular beds. In regard to the aorta, CTA can diagnose aneurysm, dissection, and wall abnormalities such as ulceration, calcification, or thrombus throughout the full length of the aorta, as well as the involvement of branch vessels. Disease of the aorta or great vessels can present with a broad clinical spectrum of symptoms and signs. The accepted diagnostic gold standard, selective digital subtraction angiography, is now being challenged by state-of-the-art CTA and magnetic resonance (MR) angiography. Currently, in many centers, cross-sectional imaging modalities are being used as the first line of diagnosis to evaluate the vascular system, and conventional angiography is reserved for therapeutic intervention.

© Springer International Publishing 2016 M.J. Budoff, J.S. Shinbane (eds.), Cardiac CT Imaging: Diagnosis of Cardiovascular Disease, DOI 10.1007/978-3-319-28219-0_18

319

320

A. Alani and M.J. Budoff

The following broad approach is a guide to CT scan acquisition for various scanners. For peripheral imaging, where electrocardiogram (ECG) gating is not required, 16–320-slice scanners are more than adequate to image the entire volume. In addition, there is no need for the speed that is required for cardiac work (temporal resolution or rotation speed).

Fig. 18.1 Axial view of pseudo-dissection of aorta caused by motion artifact in an ungated computed tomography (CT) scan of the chest

Principles of Imaging In aortic imaging, the volume coverage capabilities of MDCT have come to full use without having to compromise on resolution or detail [2, 3]. With the current configuration of 64-row (or greater) CT scanners, the entire abdominal aorta and the iliac arteries can be covered within seconds and with isotropic resolution (Chap. 1). Investigation of the dataset can now be done on the anteroposterior (coronal) and lateral (sagittal) planes, which has been the convention with invasive angiography. A few important technical advances have further improved aorta imaging using CTA. First is the increased number of detector rows for the acquisition of images over greater z-axis lengths with one gantry rotation. With up to 320 detectors, volume coverage per rotation is as much as 160 mm. The typical distance needed for the abdominal aorta is on the order of 400 mm, so two to three rotations would cover the entire abdomen. Using a rotation speed of 50 and 70 %, VR: 95 and 90 %; MIP: 86 and 68 %, respectively). The image quality obtained with VR was not significantly better than that with MIP, but vascular delineation in VR images was significantly better (Fig. 18.11). The VR technique of renal MRA enabled more accurate detection and quantification of renal artery stenosis than MIP, with significantly improved vascular delineation.

Conclusion CTA is a highly reliable technique for the detection of renal artery stenosis and for morphologic assessment. CTA can surpass conventional angiography in terms of diagnostic accuracy and reduced exposure to iodinated contrast (Fig. 18.11). In patients with renal insufficiency, color-coded duplex US or gadolinium-enhanced MRA should remain as the initial examination performed, depending on local expertise and availability. However, new warnings regarding systemic fibrosis with gadolinium make this agent contraindicated in patients with glomerular filtration rates of 1:1 in 3 or 4 lobes Retrograde opacification of the inferior vena cava or hepatic vein

pulmonary circulation due to a complex process intrinsic to the pulmonary vasculature. As many entities clinically mimic PAH, it is a diagnosis of exclusion, requiring a thorough workup and proper consideration of more common etiologies such as left-sided heart disease and hypoxic lung disease. PAH, like PH, is similarly defined hemodynamically as a mean pulmonary artery pressure ≥25 mmHg at rest and PVR >3 Wood units, but the pulmonary capillary wedge pressure measures ≤15 mmHg [3]. The natural history of PAH is variable based upon etiology but it typically follows a progressive course with a poor prognosis if left untreated [4]. Most patients with PAH present with exertional dyspnea that worsens over months to years. Exertional angina, syncope, and peripheral edema appear later in the course when increasing pulmonary vascular resistance strains and ultimately impairs right ventricular function. The diagnosis of PAH is often delayed due to the nonspecific symptoms and subtle findings on physical examination {5].

Idiopathic and Heritable PAH Idiopathic PAH (IPAH) is a diagnosis of exclusion in which no etiology nor family history can account for the disease. The incidence of IPAH is rare, with an estimated incidence of 1–2 cases per million per year worldwide [6]. The disorder is approximately 4 times more common in women [7, 8], presenting in the third decade for women and in the fourth decade for men without racial or ethnic predisposition [6]. PAH has a familial component is some patients. Germline mutations in the bone morphogenetic protein receptor type 2 (BMPR2) gene can be detected in approximately 70 % of cases [8, 9]. Mutations in activin receptor-like kinase type 1 (ALK-1 or endoglin have been found in familial PH with a strong association for concomitant hereditary hemorrhagic

B.S. Messenger and R.J. Oudiz

telangiectasia. BMPR2 mutations, however, have also been detected in 11 % to 40 % of apparently idiopathic cases without a family history [10, 11], so the distinction between idiopathic and familial BMPR2 mutations may be artificial. Interestingly, in up to 30 % of families with PAH, no BMPR2 mutation has been identified. Thus, heritable forms of PAH include IPAH with germline mutations and familial cases with or without identified germline mutations [12, 13]. Genetic testing is not mandatory in heritable PAH, genetic testing should only be performed after genetic counseling, with a discussion of the risks, benefits, and limitations of such testing [14].

PAH Associated with Connective Tissue Diseases The prevalence of PAH has been well established for patients with systemic sclerosis (SSc). Two recent prospective studies using echocardiography as a screening method and right heart catheterization for confirmation found a prevalence of PAH in SS of between 7 % and 12 % [15, 16]. The prevalence of PAH in systemic lupus erythematosis and mixed connective tissue disease remains unknown; although its incidence is greater than IPAH, it occurs less frequently than in SSc [17–20]. In the absence of fibrotic lung disease, PAH has also been reported infrequently in Sjögren syndrome [21], polymyositis [22], and rheumatoid arthritis [23]. Approximately one half of patients with PH and connective tissue diseases will die within 1 year if left untreated [24]. PH is also a frequent complication of idiopathic pulmonary fibrosis (IPF). IPF patients with concomitant PH have a two to threefold increase in mortality compared to IPF patients with normal pulmonary arterial pressures [25]. However, the presence or severity of PH does not correlate with the level of IPF disease seen on high resolution CT [26]. It is increasingly being recognized that PH in patients with IPF is the sequelae of a “primary” occlusive pulmonary vasculopathy, rather than being purely secondary to fibrotic destruction of the vascular bed [27].

PAH Associated with Congenital Heart Disease (CHD) PAH related to CHD results from the effects of a longstanding abnormal increase in pulmonary blood flow which leads to pathologic changes in the pulmonary vasculature, particularly in the smaller vessels. This results in increased pulmonary vascular resistance, which in turn has deleterious effects upon the heart and larger pulmonary vascular structures. Direct shunts can increase pulmonary blood flow and pressure (patent ductus arteriosis) (Fig. 19.1). Eisenmenger syndrome is defined as CHD with an initially large systemic-to-pulmonary shunt that induces progressive pulmonary vascular disease and PAH, resulting in reversal of the shunt and central cyanosis [28, 29]. Eisenmenger syndrome represents the most advanced form

19 Assessment of Pulmonary Vascular Disease

339

systemic-to-pulmonary shunts in Europe and North America ranges between 1.6 and 12.5 cases per million adults, with 25–50 % of this population affected by Eisenmenger syndrome [33].

PAH Associated with Schistosomiasis Though not prominent in industrialized nations, schistosomiasis may be the most common cause of PAH in the world given the expected 200 million people infected with the parasite [34]. The mechanism of PH is probably multifactorial, and includes mechanical obstruction, local vascular inflammation related to eggs, and portal hypertension [35, 36]. These cases can have a similar clinical presentation to IPAH [37], with similar histopathologic findings [38], and thus similar radiographic appearance. PAH occurs almost exclusively in the 10 % of infected patients who develop hepatosplenic schistosomiasis [39].

Fig. 19.1 64 slice multi-detector CT in a 20 year-old woman with patent ductus arteriosis (black arrow), seen well in this sagittal view with resultant dilation of the main pulmonary artery

Fig. 19.2 CT scan in a 22 year-old woman with Eisenmenger syndrome due to atrial septal defect and pulmonary hypertension. Right cardiac chambers are enlarged with marked right ventricular hypertrophy (black arrow)

of PAH associated with CHD (Fig. 19.2). A large proportion of patients with CHD develop some degree of PAH [30–32]. The prevalence of PAH associated with congenital

Pulmonary Veno-Occlusive Disease and Pulmonary Capillary Hemangiomatosis Pulmonary veno-occlusive disease (PVOD) accounts for a small number of pulmonary hypertension cases, most commonly in children and young adults [40]. The estimated annual incidence rate is 0.1–0.2 cases of PVOD per million persons in the general population [41, 42]. Patients with PH due to PVOD frequently present with congestive heart failure symptoms of dyspnea on exertion, peripheral edema, and radiographic signs of pulmonary edema with Kerley B lines that is expected with postcapillary PH. However, the hemodynamics at right heart catheterization similar to pre-capillary PH and may lead clinicians to diagnose idiopathic PH [40]. Vasodilator therapy, as used to treat a subset of idiopathic PH patients, can cause acute pulmonary edema, so distinguishing PVOD from IPAH has important clinical significance. CT imaging may help raise a clinician’s suspicion for PVOD. The CT findings of the disease includes smooth septal thickening, diffuse or mosaic ground-glass opacities, multiple small nodules, and pleural effusion [43–47]. PVOD and pulmonary capillary hemangiomatosis (PCH) also show the presence of crackles and clubbing on physical examination, hemosiderin-laden macrophages on bronchoalveolar lavage [48], as well as lower carbon monoxide diffusing capacity and PaO2 [47]. In a comparison study, PVOD patients had peripheral ground glass opacities (GGO) in 93 % of cases compared with 13 % incidence of GGO in patients with PAH [49]. Thickened interlobular septa and mediastinal adenopathy were also strongly correlated with PVOD compared with PAH patients [41]. The only effective treatment for PVOD is lung transplant with most diagnoses being made at time of transplant or autopsy.

340

Group 2 PH: Pulmonary Hypertension due to Left Heart Disease One of the most commonly seen forms of PH is pulmonary hypertension due to left heart disease. Since the etiology of their PH is distal to the pulmonary arterial system, Group 2 patients have “post-capillary” PH. Like other forms of PH, it manifests clinical signs and symptoms of the right heart failure syndrome. Group II PH includes patients with preserved or reduced ejection fraction, valvular heart disease and certain forms of congenital heart disease affecting left ventricular flow and/or performance.

Group 3 PH: Pulmonary Hypertension due to Lung Diseases and/or Hypoxia A subcategory of lung disease characterized by a mixed obstructive and restrictive pattern includes chronic bronchiectasis, cystic fibrosis [50], and a newly identified syndrome characterized by the combination of pulmonary fibrosis, mainly of the lower zones of the lung, and emphysema, mainly of the upper zones of the lung [51]. The prevalence of PH in all of these conditions remains largely unknown. However, in a recent retrospective study of 998 patients with chronic obstructive pulmonary disease who underwent right heart catheterization, only 1 % had severe pulmonary hypertension (mean PA pressure >40 mm Hg) [52]. In the syndrome of combined pulmonary fibrosis and emphysema, the prevalence of PH is almost 50 % [51].

B.S. Messenger and R.J. Oudiz

Diagnostic Workup of PE Many diagnostic tests have been suggested for the evaluation of patients with suspected VTE. These include the history and physical examination to the electrocardiogram, chest radiography, echocardiography, ventilation-perfusion scintigraphy, pulmonary angiography, CT and MR angiography, lower-extremity venography, and sonography. Although the diagnostic accuracy of laboratory tests such as D-dimer has increased (a negative result in combination with a low-probability clinical assessment provides reasonable certainty for excluding PE), radiographic imaging plays an important role in the diagnosis of PE, especially with the development of multi-detector CT (MDCT) and increased use of CT pulmonary angiography. Although normal chest x-ray findings are observed in 24 % of patients with PE, an elevated hemidiaphragm can be observed in 20 % of patients with acute PE [58]. An elevated hemidiaphragm, consolidation, pleural effusion, or atelectasis occurs in about 2/3 of patients with acute PE. Especially in a massive PE (Fig. 19.3), local hyperlucency is seen when a lobar or segmental artery is occluded (Westermark sign), and engorgement of a major hilar artery (Fleischner sign) can be detected [59, 60]. Abrupt tapering or termination of a pulmonary vessel (knuckle sign), a pleural-based density or costophrenic density (Hampton’s hump), and alveolar or interstitial pulmonary edema may occur. Most of the above chest x-ray findings are nonspecific. Nuclear scintigraphy (ventilation-perfusion or V/Q scanning) is useful if multidetector CT angiography (MDCTA) is not available. The V/Q scan in a patient with an acute PE will demonstrate an area

Group 4 PH: Chronic Thromboembolic Pulmonary Hypertension (CTEPH) Pulmonary embolism (PE) is an obstruction of a pulmonary artery caused by a blood clot, air, fat, or tumor tissue. The most common cause of the obstruction is a blood clot (thrombus) usually from a peripheral vein. The average annual incidence of venous thromboembolism (VTE) in the United States is 1 per 1000, with about 250,000 incident cases occurring annually [53–55]. The challenge in understanding the real disease is that an additional equal number of patients are diagnosed with PE at autopsy [53, 56]. It is estimated that between 650,000 and900,000 fatal and nonfatal VTE events occur in the US annually [57]. The classic triad of signs and symptoms of PE (hemoptysis, dyspnea, chest pain) are neither sensitive nor specific, and many patients with PE are initially asymptomatic; most patients who have symptoms often have atypical and/or nonspecific symptoms.

Fig. 19.3 64 slice multi-detector CT in a patient with pulmonary embolism. An axial section at the level of the main pulmonary artery shows the filling defect (white arrows) of bilaterally enlarged pulmonary arteries with massive thromo-emboli. A small amount of right pleural effusion (white arrowhead) is observed

19 Assessment of Pulmonary Vascular Disease

distal to thrombus that is not properly per fused, increasing the V/Q mismatch. Perfusion defects appear the same on a VQ scan whether the pulmonary embolus is acute or chronic. In contrast, CT findings of acute versus chronic emboli differ. In CTEPH, there may be intraluminal thrombi and dilated bronchial arteries. Furthermore, a mosaic pattern of hyperattenuation and hypoattenuation is frequently seen, representing the variation in perfusion between pulmonary segments [40]. Other evidence of CTEPH on CT includes abrubt narrowing of vessels, intimal irregularities, or total occlusions. Yet, there is greater concern that CT findings may be difficult to detect. A study from Tunariu, et al. found a sensitivity of 51 % for CTPA compared with 96 % sensitivity for VQ scans [61]. The specificity of CTPA can also be limited by the subtly distinct mosaic pattern of hyperattenuation and hypoattenuation found in Group 1 PAH. With improved ECG-gated CT imaging, recent studies have shown much improved sensitivity. He, et al found the sensitivity of VQ scan to be 100 % if high and intermediate risk results were interpreted as positive for a PE with sensitivity declining to 96 % but specificity rising to 95 % if only high risk results were counted as abnormal [62]. CT had a sensitivity of 92 % with specificity of 95 % in the same patient cohort [62]. While V/Q scan remains slightly more sensitive, CT imaging now serves as a viable alternative given improvements in sensitivity and specificity with newer scanners. Pulmonary angiography, the gold standard for diagnosing PE, is being replaced in many institutions with MDCTA, which is less invasive, easier to perform, and has high sensitivity (83 %-100 %) and specificity (89 %-97 %) [63–65]. The PIOPED studies (large multicenter trials for CCTA in suspected PE) report negative predictive values as high as 99 %. The newest noninvasive method for the evaluation and diagnosis of PE is MRI. Although not as extensively studied as other imaging techniques, it can be utilized for the patients with renal dysfunction or an iodine contrast allergy. CT angiographic findings are shown (Figs. 19.3 and 19.4). Chronic thromboembolic pulmonary hypertension (CTEPH) represents a frequent cause of PH (Fig. 19.4). The incidence of CTEPH is uncertain; however, it is known to occur in up to 4 % of patients after an acute pulmonary embolism [66, 67]. It is strongly recommended that patients with suspected or confirmed CTEPH be referred to a center with expertise in the management of this disease to consider the feasibility of performing pulmonary thromboendarterectomy, currently the only curative treatment. The decision depends on the location of the obstruction (central vs. more distal pulmonary arteries), the correlation between hemodynamic findings, and the degree of mechanical obstruction.

341

Fig. 19.4 64 slice multi-detector CT in a 58 year-old man with multiple bilateral pulmonary embolism. (a) “Saddle embolus” (black arrow) at the bifurcation of right pulmonary Artery (PA). (b) “Tram line” (white arrow) and “ring shape” (black arrowhead) features of emboli at both peripheral PAs

Methods of Detecting and Characterizing PH The appropriate classification of patients with suspicion of PAH requires a rigorous diagnostic approach. PH symptoms are nonspecific and progress slowly over months to years. Exertional dyspnea is the most common symptom with fatigue occurring in a minority of cases. Manifestations of syncope, peripheral edema, and angina may be more concerning and frequently appear in patients with evidence of impaired right ventricular function. Clinicians should have a higher index of suspicion for PAH in patients using amphetamines or diet pills, a history of autoimmune disease, or a known family history of PH. Guidelines put forth from the AHA/ACCF recommend a rigorous workup for pulmonary hypertension which includes the diagnostic algorithm as shown in Table 19.2 [3]. Transthoracic echocardiography (TTE) is often used as a first-line screening test to exclude or identify patients with severe PH by estimating systolic PA pressure and looking for evidence of the cardiac hemodynamic perturbations seen with PAH, such as right-sided cardiac and great vessel

342 Table 19.2 Clinical classification of pulmonary hypertension 1. Pulmonary arterial hypertension (PAH) 1. Idiopathic PAH 2. Heritable BMPR2 mutation (familial or isolated) ALK1, endoglin (with or without hereditary hemorrhagic telangiectasia) Unknown 3. Drug- and toxin-induced 4. Associated with Connective tissue diseases HIV infection Portal hypertension Congenital heart diseases Schistosomiasis Chronic hemolytic anemia 5. Persistent pulmonary hypertension of the newborn 6. Pulmonary veno-occlusive disease (PVOD) and/or pulmonary capillary hemangiomatosis (PCH) 2. Pulmonary hypertension due to left heart disease 1. Systolic dysfunction 2. Diastolic dysfunction 3. Valvular disease 3. Pulmonary hypertension due to lung diseases and/or hypoxia 1. Chronic obstructive pulmonary disease 2. Interstitial lung disease 3. Other pulmonary diseases with mixed restrictive and obstructive pattern 4. Sleep-disordered breathing 5. Alveolar hypoventilation disorders 6. Chronic exposure to high altitude 7. Developmental abnormalities 4. Chronic thromboembolic pulmonary hypertension (CTEPH) 5. Pulmonary hypertension with unclear multifactorial mechanisms 1. Hematologic disorders: myeloproliferative disorders, splenectomy 2. Systemic disorders: sarcoidosis, pulmonary Langerhans cell histiocytosis, lymphangioleiomyomatosis, neurofibromatosis, vasculitis 3. Metabolic disorders: glycogen storage disease, Gaucher disease, thyroid disorders 4. Others: tumoral obstruction, fibrosing mediastinitis, chronic renal failure on dialysis ALK1 activin receptor-like kinase type1, BMPR2 bone morphogenetic protein receptor type2

chamber enlargement and dysfunction, flattening of the interventricular septum, and pericardial effusion. Echocardiography is also useful for evaluating congenital heart disease and left-sided heart disease. Suggested radiographic imaging includes a chest x-ray and, depending on the need, CT radiography. The chest radiographic findings of PH are hilar fullness characteristic of dilated central pulmonary arteries, pruning of the

B.S. Messenger and R.J. Oudiz

peripheral arteries, and right-sided cardiac chamber enlargement [3, 5]. The chest radiograph may suggest an underlying cause for PH and is thus recommended in the workup of suspected PH. Right-sided heart catheterization (RHC) is the most accurate test for determining cardiopulmonary hemodynamics and remains the gold standard by which the diagnosis of PH is made and hemodynamic severity is calculated. A RHC is required not only to confirm the presence and the severity of PH, but also to exclude left-sided heart disease, potentially correctable intracardiac left-to-right shunting, and to perform acute vasodilator testing. Because the signs and symptoms of PH are non-specific and there is no reliable non-invasive test for its detection, patients often undergo computed tomography (CT) as part of their diagnostic work-up. CT is able to evaluate the lung parenchyma (for interstitial or emphysematous changes), the pulmonary artery (calcification, dilation, embolism, patent ductus), the pulmonary veins, the cardiac chambers (hypertrophy, dysplasia, enlargement, thrombus, septal defects), the coronary vessels, and the IVC simultaneously. It is important to be aware of the CT findings that may suggest the diagnosis of PH, such as an enlarged main pulmonary artery (Figs. 19.5). Radiographically, PH is said to be more likely when the main pulmonary artery diameter (MPAD) is ≥29 mm (sensitivity 69 %, specificity 100 %) [6, 7] and/or the ratio of the main pulmonary artery to ascending aorta diameter is >1 (sensitivity 70.8 % and specificity 76.5 %) [8, 68]. Others have reported that the most specific CT findings for the presence of PH were both a MPAD ≥29 mm and segmental artery-to-bronchus ratio of >1:1 in three or four lobes (specificity 100 %) [9]. In addition, the MPAD correlates with the severity of pulmonary hypertension; two studies have defined the upper limit of normal for main pulmonary artery diameter as 32 mm [6, 10]. An additional feature of PH is rapid tapering or “pruning” of the distal pulmonary vessels (Fig. 19.5). With improvements in ECG-gated CT technology, newer methods have been developed with similar accuracy in predicting the presence of PH. Revel et al. demonstrated that a reduction in distensibility of the right pulmonary artery has a high specificity for PH[69]. Distensibility is determined by evaluating the change in cross-sectional area of the right PA during systole compared with diastole. The difference in maximum area from the minimum area is divided by the maximum area and multiplied by 100 to get a percentage of distensibility. A value of less than 16.5 % had 86 % sensitivity and 96 % specificity for PH [69]. RV thickness is a known finding in PH with RV free wall thickness >6 mm (81 % sensitivity and 91 % specificity), RV/LV lumen ratio >1.28 (sensitivity of 85 % and specificity of 86 %), and RV wall/LV wall ratio >0.32 as accurate predictors of PH [70]. With a high degree of specificity, CT imaging provides important

19 Assessment of Pulmonary Vascular Disease

343

Fig. 19.6 64 slice multi-detector CT in a 44 year-old man with secondum atrial septal defect (ASD) and pulmonary hypertension demonstrating right atrial (RA) and right ventricular (RV) enlargement

Fig. 19.5 Electron Beam CT scan in a 22 year-old woman (a) and 64 slice multi-detector CT in a 40 year-old man (b), both with pulmonary hypertension. Pulmonary arteries (PA) are enlarged and tapering distally, with an increased main PA to aorta ratio

evidence for pulmonary hypertension that should prompt clinicians to pursue a workup for pulmonary hypertension. The presence of calcification in the pulmonary arteries suggests more severe disease [9], as does pericardial thickening and/or effusion [12]. Some studies have reported that hypertrophy of the bronchial artery occurs frequently in patients with idiopathic PAH (IPAH) and Eisenmenger’s syndrome [13, 14]. These studies also reported that pulmonary artery thromboses and aneurysms were common in patients with congenital heart disease (CHD) as compared to patients with IPAH, while dilation and mural calcification was seen in similar frequency in both groups [13]. It is generally accepted that in patients with chronic, unrepaired systemic-to-pulmonary shunts the CT findings can appear very similar to those found in precapillary pulmonary hypertension such as IPAH and PAH associated with connective tissue disease, portal hypertension, and HIV infection. Commonly, right (or pulmonic) ventricular hypertrophy is seen, with right ventricular enlargement and associated right atrial enlargement (Fig. 19.6).

Fig. 19.7 64 slice multi-detector CT in a 31 year-old man with sinus venosum atrial septal defect (ASD) and pulmonary hypertension demonstrating right atrial (RA) dilation and right ventricular (RV) hypertrophy, and flattening of the inter-ventricular septum (black arrow), indicating high right sided pressures

As the right ventricle enlarges, the interventricular septum becomes flattened (Fig. 19.7) and eventually convex to the left side [15–17], and thus septal flattening is a commonly noted cardiac abnormality found in patients with PH [18]. If cine-CT is performed, reduced right ventricular systolic function may also be present. The presence of retrograde opacification of the inferior vena cava or hepatic vein during contrast-enhanced CT may be a nonspecific sign of

344

B.S. Messenger and R.J. Oudiz

echocardiographic data, but cine-CT provides the necessary axial images of the LV in systole and diastole to calculate the index. Although CT imaging of the great vessels is a simple and straightforward noninvasive methodology, it has not gained widespread acceptance as a screening test for PH [1]. Table 19.1 summarizes the CT findings of PH. While CT cannot diagnose pulmonary hypertension, clinicians should be aware that a wealth of information regarding the end result of the chronic hemodynamic effect of PH upon cardiovascular anatomy. A diagnosis and its underlying etiology may be highly suggested by closely evaluating the pulmonary vasculature and cardiac anatomy on CT imaging.

References

Fig. 19.8 64 slice multi-detector CT in a 26 year-old man with anolamous pulmonary venous return and pulmonary hypertension demonstrating right atrial dilation and right ventricular hypertrophy, and dilated inferior vena cava with swirling of white contrast (black arrow), indicating severe tricuspid regurgitation

significant pulmonary hypertension and/or right ventricular dysfunction [19] (Fig. 19.8). Although hemodynamics cannot be directly measured with CT, the sequelae of chronic pulmonary hypertension can be seen with structural cardiac changes. The increased pulmonary vascular resistance that occurs with PH places greater strain on the right ventricle. Right ventricular hypertrophy (>6 mm thickness) is a common finding, and as the RV begins failing in the setting of persistently elevated afterload, an increased in RV end diastolic volume will occur [71]. RV enlargement leads to increasing RV pressure and volume, which are reflected in the bowing of the interventricular septum. CT findings of RV/LV >1 and LV septal flattening are sensitive and specific markers for PH. In small studies by Contractor et al and Lim et al, found these signs had a sensitivity of 78 %–92 % and a specificity of 100 % for echocardiographic findings of RV dysfunction. [71–73] Systolic eccentricity index (sEi) correlates well with pulmonary hypertension severity [74]. In normal conditions, the LV cavity is shaped like a circle. With increasing RV pressure and volume overload, the septal wall flattening leads to a deformed, D-shaped left ventricle. The length of the septum (D1) becomes longer than the width of the LV cavity (D2), with a ratio of D1/D2 >1 (normal score is 1) that is the eccentricity score [74]. Most studies have evaluated sEi with

1. Vonk-Noordegraaf A, Haddad F, Chin KM, Forfia PR, Kawut SM, Lumens J, Naeije R, Newman J, Oudiz RJ, Provencher S, Torbicki A, Voelkel NF, Hassoun PM. Right heart adaptation to pulmonary arterial hypertension: physiology and pathobiology. J Am Coll Cardiol. 2013;62(25 Suppl):D22–33. 2. Simonneau G, Gatzoulis MA, Adatia I. Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol. 2013;62(25_S):D34–41. 3. McLaughlin VV, Archer SL, Badesch DB, et al. ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association developed in collaboration with the American College of Chest Physicians; American Thoracic Society, Inc.; and the Pulmonary Hypertension Association. J Am Coll Cardiol. 2009;53:1573–619. 4. McLaughlin VV, Presberg KW, Doyle RL. Prognosis of pulmonary arterial hypertension: ACCP evidence-based Clinical Practice Guidelines. Chest. 2004;126(1_suppl):78S–92. 5. Brown LM, Chen H, Halpern S, Taichman D, McGoon MD, Farber HW, Frost AE, Liou TG, Turner M, Feldkircher K, Miller DP, Elliott CG. Delay in recognition of pulmonary arterial hypertension: factors identified from the REVEAL Registry. Chest. 2011;140:19–26. 6. Rich S, Dantzker DR, Ayres SM, et al. Primary pulmonary hypertension: a national prospective study. Ann Intern Med. 1987; 107:216–23. 7. Loyd JE, Butler MG, Foroud TM, Conneally PM, Phillips JA, Newman JH. Genetic anticipation and abnormal gender ratio at birth in familial primary pulmonary hypertension. Am J Respir Crit Care Med. 1995;152:93–7. 8. Cogan JD, Pauciulo MW, Batchman AP, et al. High frequency of BMPR2 exonic deletions/duplications in familial pulmonary arterial hypertension. Am J Respir Crit Care Med. 2006;174:590–8. 9. Aldred MA, Vijayakrishnan J, James V, et al. BMPR2 gene rearrangements account for a significant proportion of mutations in familial and idiopathic pulmonary arterial hypertension. Hum Mutat. 2006;27:212–3. 10. Machado RD, Aldred MA, James V, et al. Mutations of the TGF- β type II receptor BMPR2 in pulmonary arterial hypertension. Hum Mutat. 2006;27:121–32. 11. Thomson JR, Machado RD, Pauciulo MW, et al. Sporadic primary pulmonary hypertension is associated with germline mutations of the gene encoding BMPR-II, a receptor member of the TGF-β family. J Med Genet. 2000;37:741–5.

19 Assessment of Pulmonary Vascular Disease 12. Chaouat A, Coulet F, Favre C, et al. Endoglin germline mutation in a patient with hereditary haemorrhagic telangiectasia and dexfenfluramine associated pulmonary arterial hypertension. Thorax. 2004;59:446–8. 13. Trembath RC, Thomson JR, Machado RD, et al. Clinical and molecular genetic features of pulmonary hypertension in patients with hereditary hemorrhagic telangiectasia. N Engl J Med. 2001;345:325–34. 14. McGoon M, Gutterman D, Steen V, et al. Screening, early detection, and diagnosis of pulmonary arterial hypertension: ACCP evidencebased clinical practice guidelines. Chest. 2004;126:14S–34. 15. Hachulla E, Gressin V, Guillevin L, et al. Early detection of pulmonary arterial hypertension in systemic sclerosis: a french nationwide prospective multicenter study. Arthritis Rheum. 2005;52:3792–800. 16. Mukerjee D, St George D, Coleiro B, et al. Prevalence and outcome in systemic sclerosis associated pulmonary arterial hypertension: application of a registry approach. Ann Rheum Dis. 2003; 62:1088–93. 17. Tanaka E, Harigai M, Tanaka M, Kawaguchi Y, Hara M, Kamatani N. Pulmonary hypertension in systemic lupus erythematosus: evaluation of clinical characteristics and response to immunosuppressive treatment. J Rheumatol. 2002;29:282–7. 18. Asherson RA, Higenbottam TW, Dinh Xuan AT, Khamashta MA, Hughes GR. Pulmonary hypertension in a lupus clinic: experience with twenty-four patients. J Rheumatol. 1990;17:1292–8. 19. Burdt MA, Hoffman RW, Deutscher SL, Wang GS, Johnson JC, Sharp GC. Long-term outcome in mixed connective tissue disease: longitudinal clinical and serologic findings. Arthritis Rheum. 1999;42:899–909. 20. Jaïs X, Launay D, Yaici A, et al. Immunosuppressive therapy in lupus- and mixed connective tissue disease-associated pulmonary arterial hypertension: a retrospective analysis of twenty-three cases. Arthritis Rheum. 2008;58:521–31. 21. Launay D, Hachulla E, Hatron PY, Jaïs X, Simonneau G, Humbert M. Pulmonary arterial hypertension: a rare complication of primary Sjögren syndrome: report of 9 new cases and review of the literature. Medicine (Baltimore). 2007;86:299–315. 22. Bunch TW, Tancredi RG, Lie JT. Pulmonary hypertension in polymyositis. Chest. 1981;79:105–7. 23. Dawson JK, Goodson NG, Graham DR, Lynch MP. Raised pulmonary artery pressures measured with doppler echocardiography in rheumatoid arthritis patients. Rheumatology (Oxford). 2000;39:1320–5. 24. Coghlan JG, Handler C. Connective tissue associated pulmonary arterial hypertension. Lupus. 2006;15:138–42. 25. Lettieri CJ, Nathan SD, Barnett SD, et al. Prevalence and outcomes of pulmonary arterial hypertension in advanced idiopathic pulmonary fibrosis. Chest. 2006;129:746–52. 26. Corte TJ, et al. Pulmonary hypertension in idiopathic pulmonary fibrosis: a review. Sarcoidosis Vasc Diffuse Lung Dis. 2009; 26:7–19. 27. Zisman DA, Karlamangla AS, Ross DJ, et al. High-resolution chest CT findings do not predict the presence of pulmonary hypertension in advanced idiopathic pulmonary fibrosis. Chest. 2007; 132:773–9. 28. Eisenmenger V. Die angeborene defecte der kammersheidewand das herzen. Z Klin Med. 1897;132:131. 29. Wood P. The Eisenmenger syndrome or pulmonary hypertension with reversed central shunt. Br Med J. 1958;2:701–9. 30. Daliento L, Somerville J, Presbitero P, et al. Eisenmenger syndrome: factors relating to deterioration and death. Eur Heart J. 1998;19:1845–55. 31. Besterman E. Atrial septal defect with pulmonary hypertension. Br Heart J. 1961;23:587–98.

345 32. Hoffman JI, Rudolph AM. The natural history of ventricular septal defects in infancy. Am J Cardiol. 1965;16:634–53. 33. Galiè N, Manes A, Palazzini M, et al. Management of pulmonary arterial hypertension associated with congenital systemic-to pulmonary shunts and Eisenmenger’s syndrome. Drugs. 2008;68:1049–66. 34. Fernandes C, Jardim C, Honanian A, Hoette S, Morinaga LK, Souza R. Schistosomiasis and Pulmonary Hypertension. Pulmonary Vascular Disorders. Prog Respir Res. Basel. Karger. 2012;41:143–8. 35. Shaw AP, Ghareed A. The pathogenesis of pulmonary schistosomiasis in Egypt with special reference to Ayerza’s disease. J Pathol Bacteriol. 1938;46:401–24. 36. de Cleva R, Herman P, Pugliese V, et al. Prevalence of pulmonary hypertension in patients with hepatosplenic mansonic schistosomiasis- prospective study. Hepatogastroenterology. 2003;50: 2028–30. 37. Lapa MS, Ferreira EV, Jardim C, Martins Bdo C, Arakaki JS, Souza R. Clinical characteristics of pulmonary hypertension patients in two reference centers in the city of Sao Paulo. Rev Assoc Med Bras. 2006;52:139–43. 38. Chaves E. The pathology of the arterial pulmonary vasculature in Manson’s schistosomiasis. Chest. 1966;50:72–7. 39. Papamatheakis DG, Mocumbi AOH, Kim NH, Mandel J. Schistosomiasis-associated pulmonary hypertension. Pulmonary Circulation. 2014;4:596–611. 40. Grosse C, Grosse A. CT findings in diseases associated with pulmonary hypertension: a current review. Radiographics. 2010;30(7): 1753–77. 41. Mandel J, Mark EJ, Hales CA. Pulmonary veno-occlusive disease. Am J Respir Crit Care Med. 2000;162:1964–73. 42. Frazier AA, Franks TJ, Mohammed TL, Ozbudak IH, Galvin JR. From the archives of the AFIP: pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis. Radiographics. 2007;27:867–82. 43. Holcomb Jr BW, Loyd JE, Ely EW, Johnson J, Robbins IM. Pulmonary veno-occlusive disease: a case series and new observations. Chest. 2000;118:1671–9. 44. Resten A, Maitre S, Humbert M, et al. Pulmonary hypertension: CT of the chest in pulmonary veno-occlusive disease. Am J Roentgenol. 2004;183:65–70. 45. Lantuéjoul S, Sheppard MN, Corrin B, Burke MM, Nicholson AG. Pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis: a clinicopathologic study of 35 cases. Am J Surg Pathol. 2006;30:850–7. 46. Ozsoyoglu AA, Swartz J, Farver CF, Mohammed TL. High resolution computed tomographic imaging and pathologic features of pulmonary veno-occlusive disease: a review of three patients. Curr Probl Diagn Radiol. 2006;35:219–23. 47. Montani D, Achouh L, Dorfmuller P. Pulmonary venoocclusive disease: clinical, functional, radiologic, and hemodynamic characteristics and outcome of 24 cases confirmed by histology. Medicine(Baltimore). 2008;87:220–33. 48. Rabiller A, Jaïs X, Hamid A, et al. Occult alveolar haemorrhage in pulmonary veno-occlusive disease. Eur Respir J. 2006;27:108–13. 49. Resten A, Maitre S, Humbert M, Rabiller A, Sitbon O, Capron F, Simonneau G, Musset D. Pulmonary hypertension: CT of the chest in pulmonary venoocclusive disease. AJR Am J Roentgenol. 2004;183:65–70. 50. Fraser KL, Tullis DE, Sasson Z, Hyland RH, Thornley KS, Hanly PJ. Pulmonary hypertension and cardiac function in adult cystic fibrosis: role of hypoxemia. Chest. 1999;115:1321–8. 51. Cottin V, Nunes H, Brillet PY, et al. Combined pulmonary fibrosis and emphysema: a distinct underrecognised entity. Eur Respir J. 2005;26:586–93.

346 52. Chaouat A, Bugnet AS, Kadaoui N, et al. Severe pulmonary hypertension and chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2005;172:189–94. 53. Silverstein MD, Heit JA, Mohr DN, Petterson TM, O'Fallon WM, Melton 3rd LJ. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population-based study. Arch Intern Med. 1998;158:585–93. 54. Tapson VF. Acute pulmonary embolism. N Engl J Med. 2008;358:1037–52. 55. Heit JA. The epidemiology of venous thromboembolism in the community. Arterioscler Thromb Vasc Biol. 2008;28:370–2. 56. Sandler DA, Martin JF. Autopsy proven pulmonary embolism in hospital patients: are we detecting enough deep vein thrombosis? J R Soc Med. 1989;82:203–5. 57. Kumar A, Ouriel K. Handbook of endovascular interventions. New York: Springer Science & Business Media; 2012. p. 404. 58. Elliott CG. Chest radiographs in acute pulmonary embolism: results from the international cooperative pulmonary embolism registry. Chest. 2000;118:33–8. 59. Stein PD, Terrin ML, Hales CA, Palevsky HI, Saltzman HA, Thompson BT, Weg JG. Clinical, laboratory, roentgenographic and electrocardiographic findings in patients with acute pulmonary embolism and no pre-existing cardiac or pulmonary disease. Chest. 1991;100:598–603. 60. Stein PD, Willis III PW, DeMets DL, Greenspan RH. Plain chest roentgenogram in patients with acute pulmonary embolism and no pre-existing cardiac or pulmonary disease. Am J Noninvas Cardiol. 1987;1:171–6. 61. Tunariu N, Gibbs SJ, Win Z, Gin-Sing W, Graham A, Gishen P, Al-Nahhas A. Ventilation-perfusion scintigraphy is more sensitive than multidetector CTPA in detecting chronic thromboembolic pulmonary disease as a treatable cause of pulmonary hypertension. J Nucl Med. 2007;48:680–4. 62. He J, Fang W, Lv B, et al. Diagnosis of chronic thromboembolic pulmonary hypertension: comparison of ventilator/perfusion scanning and multidetector computed tomography pulmonary angiography with pulmonary angiography. Nucl Med Commun. 2012;33:459–63. 63. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med. 2006;354:2317–27.

B.S. Messenger and R.J. Oudiz 64. Qanadli SD, Hajjam ME, Mesurolle B. Pulmonary embolism detection. Prospective evaluation of dual-section helical CT versus selective pulmonary arteriography in 157 patients. Radiology. 2000;217:447–55. 65. Winer-Muram HT, Rydberg J, Johnson MS, et al. Suspected acute pulmonary embolism: evaluation with multi-detector row CT versus digital subtraction pulmonary arteriography. Radiology. 2004;233:806–15. 66. Tapson VF, Humbert M. Incidence and prevalence of chronic thromboembolic pulmonary hypertension: from acute to chronic pulmonary embolism. Proc Am Thorac Soc. 2006;3:564–7. 67. Pengo V, Lensing AW, Prins MH, et al. Incidence of chronic thromboembolic pulmonary hypertension after pulmonary embolism. N Engl J Med. 2004;350:2257–64. 68. Wells JM, Dransfield MT. Pathophysiology and clinical implications of pulmonary arterial enlargement in COPD. Int J Chron Obstruct Pulmon Dis. 2013;8:509–21. 69. Revel MP, Faivre JB, Remy-Jardin M, Delannoy-Deken V, Duhamel A, Remy J. Pulmonary hypertension: ECG-gated 64-section CT angiographic evaluation of new functional parameters as diagnostic criteria. Radiology. 2009;250:558–66. 70. Chan AL, Juarez MM, Shelton DK, MacDonald T, Li CS, Lin C, Albertson TE. Novel computed tomographic chest metrics to detect pulmonary hypertension. BMC Med Imaging. 2011;11:7. 71. Ghaye B, Ghuysen A, Bruyere PJ, D'Orio V, Dondelinger RF. Can CT pulmonary angiography allow assessment of severity and prognosis in patients presenting with pulmonary embolism? What the radiologist needs to know. Radiographics. 2006;26:23–39. 72. Contractor S, Maldjian PD, Sharma VK, Gor DM. Role of helical CT in detecting right ventricular dysfunction secondary to acute pulmonary embolism. J Comput Assist Tomogr. 2002;26:587–91. 73. Lim KE, Chan CY, Chu PH, Hsu YY, Hsu WC. Right ventricular dysfunction secondary to acute massive pulmonary embolism detected by helical computed tomography pulmonary angiography. Clin Imaging. 2005;29:16–21. 74. López-Candales A, Bazaz R, Edelman K, Gulyasy B. Apical systolic eccentricity index: a better marker of right ventricular compromise in pulmonary hypertension. Echocardiography. 2010; 27:534–8.

Part V Multidisciplinary Topics

Value Based Imaging for Coronary Artery Disease: Implications for Nuclear Cardiology and Cardiac CT

20

Daniel S. Berman, Alan Rozanski, Piotr Slomka, Rine Nakanishi, Damini Dey, John D. Friedman, Sean W. Hayes, Louise E.J. Thomson, Reza Arsanjani, Rory Hachamovitch, James K. Min, Leslee J. Shaw, and Guido Germano

Abstract

Technology in cardiac computed tomography (CT) and nuclear cardiology is constantly improving. In single photon emission CT (SPECT), new cameras, reconstruction methods, and protocols have dramatically reduced radiation doses to patients. In positron emission tomography (PET), application of quantitative measurements of myocardial perfusion reserve is improving assessment of prognosis. PET/CT is routinely performed in conjunction with coronary artery calcium (CAC) scanning in many centers, extending the ability of myocardial perfusion imaging (MPI) studies to impact patient management. In cardiac CT, marked improvements in equipment and reconstruction software have also dramatically reduced the patient radiation associated with cardiac testing, and have reduced the frequency of non-diagnostic studies. New methods for combining anatomic and functional assessment with CT—CT perfusion and FFRCT measurements—are beginning to be used clinically. With the expanding capabilities of each technology, their opportunities to provide value increases. Given the changing reimbursement paradigm from a volume-based to a value-based system, the applications of each technology that will survive are those that improve relationship between outcomes and costs. With respect to coronary artery disease (CAD), a growing body of evidence exists regarding the value of specific tests in the various clinical settings in which CAD is considered. For prevention, data is strong in that CAC scanning can provide value by improving outcomes. In the patient with acute chest pain, CCTA appears to be able to shorten time in the hospital and reduce costs. In patients with

D.S. Berman, MD (*) • P. Slomka, PhD • D. Dey, PhD J.D. Friedman, MD • S.W. Hayes, MD • L.E.J. Thomson, MBChB R. Arsanjani, MD • G. Germano, PhD Departments of Imaging and Medicine, Cedars-Sinai Medical Center and the Cedars-Sinai Heart Institute, Los Angeles, CA, USA e-mail: [email protected] A. Rozanski, MD Division of Cardiology, Mt. Sinai Saint Luke’s and Roosevelt Hospitals, New York, NY, USA R. Nakanishi, MD, PhD Department of Medicine, Los Angeles Biomedical Research Institute at Harbor-UCLA, Torrance, CA, USA

R. Hachamovitch, MD Department of Nuclear Medicine, Cleveland Clinic, Heart and Vascular Institute, Cleveland, OH, USA J.K. Min, MD, FACC Department of Radiology, Dalio Institute of Cardiovascular Imaging, Weill Cornell Medical College and the NewYork Presbyterian Hospital, New York, NY, USA L.J. Shaw, PhD Department of Medicine, Emory Clinical Cardiovascular Research Institute, Emory University School of Medicine, Atlanta, GA, USA

© Springer International Publishing 2016 M.J. Budoff, J.S. Shinbane (eds.), Cardiac CT Imaging: Diagnosis of Cardiovascular Disease, DOI 10.1007/978-3-319-28219-0_20

349

350

D.S. Berman et al.

suspected stable ischemic heart disease and an intermediate pre-test likelihood of CAD, the use of CCTA appears to be valuable. In patients who have known CAD or in whom a nondiagnostic CCTA is likely, improvement in outcomes based on CCTA is less likely and testing for ischemia may be preferred. In patients with a very high likelihood of CAD or known CAD, registry data suggests that ischemia testing, such as that provided by SPECTor PET-MPI studies, may improve outcomes by improving selection of patients for revascularization. The ISCHEMIA trial will test whether a strategy basing decisions for revascularization on noninvasive assessment of ischemia improves outcomes. Test selection is highly dependent on accurate pretest risk assessment. An updated method for assessment of pre-test risk has developed which may lead to improved utilization of cardiac imaging procedures. In all of the applications of noninvasive imaging, value can only be achieved if the appropriate patients are selected for testing and if the test result changes management, such that outcomes can be improved or costs reduced. Keywords

SPECT-MPI • PET-MPI • Myocardial perfusion imaging • Single photon emission computed tomography • Positron emission tomography • Cardiac CT • Coronary CT angiography • Coronary artery calcium scanning

Introduction The rapid evolution of new medical technologies and therapies and the increasing numbers of patients being studied or treated based on these developments are leading to an unsustainable increase in health care costs. In the United States, through legislation such as the Affordable Care Act, a transition from a volume-based health care reimbursement to a value-based reimbursement. Over the past few decades, there has been a rise in the use of imaging that parallels that of the overall rise in healthcare costs. While efforts such as implementation of Appropriate Use Criteria have slowed the rise in these expenses, a continued increase in overall imaging costs is still occurring. It is virtually inevitable that value-based medical care reimbursement will be increasingly the norm and will apply to most of the use of medical imaging. What is value in imaging? Inherently, as with any commodity, value is a function of quality and cost (Fig. 20.1). In cardiology, quality ultimately implies improvement in patient outcomes. In the patient with coronary artery disease (CAD), these outcomes might reduce cardiac events such as death or myocardial infarction or improvement in quality of life. Other measures of quality include accuracy of diagnosis, efficiency of service for the patient, and reduction of any harm that might be associated with the care such as radiation, or complications from unnecessary invasive diagnostic or therapeutic procedures. Value is inversely related to costs, which are not only the costs of the imaging studies themselves, but all of the costs related

to the study, including cost increases due to downstream testing and therapies and cost decreases due to more efficient care from reduced unnecessary downstream testing and therapies. In the future, whether the payer is the government, insurance companies, or the patients themselves, it is likely that only those approaches that provide value will be purchased. In imaging, this implies an increasing penetrance of value-based imaging, with growth in testing in areas of proven value and reduction of testing in areas in which value has not been shown. In this chapter, we review the technologic developments in the nuclear cardiology and cardiac CT in light of advances that are likely to improve value and then to explore the potential “value proposition” of these modalities in the various clinical settings of suspected or known CAD in which they are applied.

Fig. 20.1 Value-based imaging: concepts

20

Value Based Imaging for Coronary Artery Disease: Implications for Nuclear Cardiology and Cardiac CT

Technologic Developments Nuclear Cardiology

351

Added value of machine learning combining supine and prone SPECT-MPI and clinical data

a

Entire population: (N = 1181) 1.0 0.9 0.8 Area

Sensitivity

0.7 0.6 0.5

ML: Quantitative + Clinical

0.94 ± 0.01**

ML: Quantitative Only

0.90 ± 0.01*

TPD

0.88 ± 0.01

0.4 0.3 0.2

*Better than TPD (p